id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-12-13 00:00:00
| authors_parsed
listlengths 1
3.44k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2512.02634
|
Yuzhu Duan
|
Yuzhu Duan, Ziwen Yang, Xiaoming Duan, Shanying Zhu
|
A Communication-Efficient Distributed Optimization Algorithm with Coupled Constraints
| null | null | null | null |
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper designs a communication-efficient distributed optimization algorithm for optimization problems subject to coupled equality constraints. By means of duality theory, the original problem is reformulated to tackle the coupled equality constraints. Furthermore, compressed communication is employed to enhance efficiency whereas introducing compression errors that degrade the performance. To address this, differential compression techniques with dynamic scaling factors are incorporated into the algorithm design. It is shown that the proposed distributed compressed algorithm achieves linear convergence under different compressors. Numerical results further demonstrate its robust performance under different types of compressors while satisfying the equality constraints.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 10:46:29 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:58:42 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Duan",
"Yuzhu",
""
],
[
"Yang",
"Ziwen",
""
],
[
"Duan",
"Xiaoming",
""
],
[
"Zhu",
"Shanying",
""
]
] |
2512.02637
|
Hao Chu
|
Hao Chu, Haotian Zhang, Zhili Zhang
|
Higgs physics in superconductors
| null |
Acta Phys. Sin., 2025, 74(11): 117402
|
10.7498/aps.74.20250241
| null |
cond-mat.supr-con cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As pointed out by Nambu&Goldstone, continuous symmetry breaking gives rise to gapless bosonic excitation. In superconductors, continuous local U(1) gauge symmetry is broken. The gapless excitation thus created is the collective phase mode of a superconductor. In 1962, Anderson pointed out that Coulomb interaction lifts this gapless mode to the superconducting plasma frequency. Therefore, in a superconducting fluid there are no bosonic excitations below the Cooper pair binding energy. Anderson mechanism also implies that the massless photon becomes massive in a superconductor. It provides a microscopic theory for dissipationless charge transport (with Landau criterion for superfluidity) and the Meissner effect in a superconductor. Jumping to particle physics, to explain why the gauge bosons for electroweak interaction are massive, Higgs, Englert, Kibble et al. proposed the existence of the Higgs field. This matter field couples to the massless W, Z bosons and generates mass via the Higgs mechanism. Due to their conceptual similarities, these two mechanisms are referred to as the Anderson-Higgs mechanism. In 2013, the detection of the Higgs boson provided the final proof for the Higgs hypothesis decades after its conception. The amplitude mode of a superconductor, which corresponds to the Higgs boson in the above analogy, is referred to as the Higgs mode. Its spectroscopic detection has also remained elusive for decades. In recent years, the development of ultrafast techniques enabled an effective approach for studying the mode. Here, we introduce the Higgs mode from a perspective of why superconductors superconduct and review the recent development in Higgs spectroscopy, particularly in the field of nonlinear terahertz spectroscopy. We discuss the novel insights that may be learnt from these studies for future high-Tc superconductivity and correlated materials research in general.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 10:51:13 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:31:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chu",
"Hao",
""
],
[
"Zhang",
"Haotian",
""
],
[
"Zhang",
"Zhili",
""
]
] |
2512.02734
|
Liqun Qi
|
Yi Xu, Chunfeng Cui and Liqun Qi
|
Sum of Squares Decompositions for Structured Biquadratic Forms
| null | null | null | null |
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper studies sum-of-squares (SOS) representations for structured biquadratic forms. We prove that diagonally dominated symmetric biquadratic tensors are always SOS. For the special case of symmetric biquadratic forms, we establish necessary and sufficient conditions for positive semi-definiteness of monic symmetric biquadratic forms, characterize the geometry of the corresponding PSD cone as a convex polyhedron, and prove that every such PSD form is SOS for any dimensions $m$ and $n$. We also formulate conjectures regarding SOS representations for symmetric M-biquadratic tensors and symmetric $\mathrm{B}_{0}$-biquadratic tensors, discussing their likelihood and potential proof strategies. Our results advance the understanding of when positive semi-definiteness implies sum-of-squares decompositions for structured biquadratic forms.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 13:16:17 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:52:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Yi",
""
],
[
"Cui",
"Chunfeng",
""
],
[
"Qi",
"Liqun",
""
]
] |
2512.02755
|
Charles Creffield
|
Charles Creffield, Fernando Sols, Marco Schir\`o and Nathan Goldman
|
Reaching Sachdev-Ye-Kitaev physics by shaking the Hubbard model
| null | null | null | null |
cond-mat.quant-gas cond-mat.mes-hall cond-mat.str-el gr-qc quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The Sachdev-Ye-Kitaev (SYK) model has attracted widespread attention due to its relevance to diverse areas of physics, such as high temperature superconductivity, black holes, and quantum chaos. The model is, however, extremely challenging to realize experimentally. In this work, we show how a particular form of Floquet engineering, termed ``kinetic driving'', effectively eliminates single-particle processes and creates quasi-random all-to-all interactions when applied to models of Hubbard type. For the specific case of the Bose-Hubbard model, we explicitly verify that the driven system indeed reproduces SYK physics by direct comparison of the spectral form factor and out-of-time ordered correlation functions (OTOCs). Our findings indicate that a cold-atom realization of kinetic driving -- achieved through modulation of hopping amplitudes in an optical lattice -- offers a practical and accurate platform for quantum simulation of the SYK model.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 13:35:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Creffield",
"Charles",
""
],
[
"Sols",
"Fernando",
""
],
[
"Schirò",
"Marco",
""
],
[
"Goldman",
"Nathan",
""
]
] |
2512.02885
|
Nataliia Kopteva
|
Vladimir L. Zhiliakov, Nataliia E. Kopteva, Irina A. Yugova, Dmitri R. Yakovlev, Ilya A. Akimov, and Manfred Bayer
|
Exciton spin structure in lead halide perovskite semiconductors explored via the spin dynamics in magnetic field
| null | null | null | null |
cond-mat.other physics.optics
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We theoretically investigate the spin structure and spin dynamics of excitons in bulk lead halide perovskite semiconductors with cubic, tetragonal, and orthorhombic crystal symmetry. The exciton spin structure and its modification by an external magnetic field are modeled for different regimes defined by the relative magnitude of the electron-hole exchange interaction (splitting between dark and bright states) and the Zeeman spin splitting. The effects of crystal symmetry and magnetic field orientation with respect to the crystal axes are considered for lead halide perovskite crystals with band gaps in the range 1.4 - 3.5 eV, having different ratios of electron and hole g-factors. For cubic symmetry, in a longitudinal magnetic field, our theory predicts quantum beats between the bright exciton states under linearly polarized excitation and detection, while the dark exciton remains optically inactive. In a transverse magnetic field, all exciton spin states become optically active and can be excited by circularly polarized light. Reduction of the crystal symmetry leads to a zero-field offset of the exciton Larmor precession frequencies, modifying the Zeeman splitting energy dependence on magnetic field. This theoretical framework allows for the extraction of the strength of the exchange interaction and the crystal symmetry. Experimentally, we measure the exciton spin coherence via time-resolved photoluminescence at a temperature of 1.6 K in longitudinal and transverse magnetic fields in orthorhombic MAPbI3 crystals. Polarization beats at the frequency of the bright exciton are observed in both configurations. Comparison with theory indicates that the excitons are in the strong exchange interaction regime, and the reduction of symmetry does not lead to a significant splitting of the exciton spin levels.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 15:44:31 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:27:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhiliakov",
"Vladimir L.",
""
],
[
"Kopteva",
"Nataliia E.",
""
],
[
"Yugova",
"Irina A.",
""
],
[
"Yakovlev",
"Dmitri R.",
""
],
[
"Akimov",
"Ilya A.",
""
],
[
"Bayer",
"Manfred",
""
]
] |
2512.02899
|
Zhuobai Dong
|
Zhuobai Dong, Rui Zhao, Songjie Wu, Junchao Yi, Linjie Li, Zhengyuan Yang, Lijuan Wang, Alex Jinpeng Wang
|
Glance: Accelerating Diffusion Models with 1 Sample
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Diffusion models have achieved remarkable success in image generation, yet their deployment remains constrained by the heavy computational cost and the need for numerous inference steps. Previous efforts on fewer-step distillation attempt to skip redundant steps by training compact student models, yet they often suffer from heavy retraining costs and degraded generalization. In this work, we take a different perspective: we accelerate smartly, not evenly, applying smaller speedups to early semantic stages and larger ones to later redundant phases. We instantiate this phase-aware strategy with two experts that specialize in slow and fast denoising phases. Surprisingly, instead of investing massive effort in retraining student models, we find that simply equipping the base model with lightweight LoRA adapters achieves both efficient acceleration and strong generalization. We refer to these two adapters as Slow-LoRA and Fast-LoRA. Through extensive experiments, our method achieves up to 5 acceleration over the base model while maintaining comparable visual quality across diverse benchmarks. Remarkably, the LoRA experts are trained with only 1 samples on a single V100 within one hour, yet the resulting models generalize strongly on unseen prompts.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 16:05:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 11:53:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dong",
"Zhuobai",
""
],
[
"Zhao",
"Rui",
""
],
[
"Wu",
"Songjie",
""
],
[
"Yi",
"Junchao",
""
],
[
"Li",
"Linjie",
""
],
[
"Yang",
"Zhengyuan",
""
],
[
"Wang",
"Lijuan",
""
],
[
"Wang",
"Alex Jinpeng",
""
]
] |
2512.03040
|
Zeqi Xiao
|
Zeqi Xiao, Yiwei Zhao, Lingxiao Li, Yushi Lan, Ning Yu, Rahul Garg, Roshni Cooper, Mohammad H. Taghavi, Xingang Pan
|
Video4Spatial: Towards Visuospatial Intelligence with Context-Guided Video Generation
|
Project page at https://xizaoqu.github.io/video4spatial/
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We investigate whether video generative models can exhibit visuospatial intelligence, a capability central to human cognition, using only visual data. To this end, we present Video4Spatial, a framework showing that video diffusion models conditioned solely on video-based scene context can perform complex spatial tasks. We validate on two tasks: scene navigation - following camera-pose instructions while remaining consistent with 3D geometry of the scene, and object grounding - which requires semantic localization, instruction following, and planning. Both tasks use video-only inputs, without auxiliary modalities such as depth or poses. With simple yet effective design choices in the framework and data curation, Video4Spatial demonstrates strong spatial understanding from video context: it plans navigation and grounds target objects end-to-end, follows camera-pose instructions while maintaining spatial consistency, and generalizes to long contexts and out-of-domain environments. Taken together, these results advance video generative models toward general visuospatial reasoning.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 18:59:44 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:18:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xiao",
"Zeqi",
""
],
[
"Zhao",
"Yiwei",
""
],
[
"Li",
"Lingxiao",
""
],
[
"Lan",
"Yushi",
""
],
[
"Yu",
"Ning",
""
],
[
"Garg",
"Rahul",
""
],
[
"Cooper",
"Roshni",
""
],
[
"Taghavi",
"Mohammad H.",
""
],
[
"Pan",
"Xingang",
""
]
] |
2512.03346
|
Lynn Kandakji
|
Lynn Kandakji, William Woof, Nikolas Pontikos
|
Hierarchical Attention for Sparse Volumetric Anomaly Detection in Subclinical Keratoconus
|
16 pages, 7 figures, 6 tables
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
The detection of weak, spatially distributed anomalies in volumetric medical imaging remains challenging due to the difficulty of integrating subtle signals across non-adjacent regions. This study presents a controlled comparison of sixteen architectures spanning convolutional, hybrid, and transformer families for subclinical keratoconus detection from three-dimensional anterior segment optical coherence tomography (AS-OCT). The results demonstrate that hierarchical architectures achieve 21-23% higher sensitivity and specificity, particularly in the difficult subclinical regime, outperforming both convolutional neural networks (CNNs) and global-attention Vision Transformer (ViT) baselines. Mechanistic analyses indicate that this advantage arises from spatial scale alignment: hierarchical windowing produces effective receptive fields matched to the intermediate extent of subclinical abnormalities, avoiding the excessive locality observed in convolutional models and the diffuse integration characteristic of pure global attention. Attention-distance measurements show that subclinical cases require longer spatial integration than healthy or overtly pathological volumes, with hierarchical models exhibiting lower variance and more anatomically coherent focus. Representational similarity further indicates that hierarchical attention learns a distinct feature space that balances local structure sensitivity with flexible long-range interactions. Auxiliary age and sex prediction tasks demonstrate moderately high cross-task consistency, supporting the generalizability of these inductive principles. The findings provide design guidance for volumetric anomaly detection and highlight hierarchical attention as a principled approach for early pathological change analysis in medical imaging.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 01:20:13 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:30:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kandakji",
"Lynn",
""
],
[
"Woof",
"William",
""
],
[
"Pontikos",
"Nikolas",
""
]
] |
2512.03420
|
Kang Yang
|
Kang Yang, Yunhang Zhang, Zichuan Li, Guanhong Tao, Jun Xu, Xiaojing Liao
|
HarnessAgent: Scaling Automatic Fuzzing Harness Construction with Tool-Augmented LLM Pipelines
| null | null | null | null |
cs.CR cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language model (LLM)-based techniques have achieved notable progress in generating harnesses for program fuzzing. However, applying them to arbitrary functions (especially internal functions) \textit{at scale} remains challenging due to the requirement of sophisticated contextual information, such as specification, dependencies, and usage examples. State-of-the-art methods heavily rely on static or incomplete context provisioning, causing failure of generating functional harnesses. Furthermore, LLMs tend to exploit harness validation metrics, producing plausible yet logically useless code. % Therefore, harness generation across large and diverse projects continues to face challenges in reliable compilation, robust code retrieval, and comprehensive validation.
To address these challenges, we present HarnessAgent, a tool-augmented agentic framework that achieves fully automated, scalable harness construction over hundreds of OSS-Fuzz targets. HarnessAgent introduces three key innovations: 1) a rule-based strategy to identify and minimize various compilation errors; 2) a hybrid tool pool for precise and robust symbol source code retrieval; and 3) an enhanced harness validation pipeline that detects fake definitions. We evaluate HarnessAgent on 243 target functions from OSS-Fuzz projects (65 C projects and 178 C++ projects). It improves the three-shot success rate by approximately 20\% compared to state-of-the-art techniques, reaching 87\% for C and 81\% for C++. Our one-hour fuzzing results show that more than 75\% of the harnesses generated by HarnessAgent increase the target function coverage, surpassing the baselines by over 10\%. In addition, the hybrid tool-pool system of HarnessAgent achieves a response rate of over 90\% for source code retrieval, outperforming Fuzz Introspector by more than 30\%.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 03:55:09 GMT"
},
{
"version": "v2",
"created": "Fri, 05 Dec 2025 21:12:52 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 04:13:33 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yang",
"Kang",
""
],
[
"Zhang",
"Yunhang",
""
],
[
"Li",
"Zichuan",
""
],
[
"Tao",
"Guanhong",
""
],
[
"Xu",
"Jun",
""
],
[
"Liao",
"Xiaojing",
""
]
] |
2512.03432
|
Jose Cruz
|
Jos\'e Cruz
|
Classification of totally real number fields via their zeta function, regulator, and log unit lattice
|
24 pages. Comments are welcome!
| null | null | null |
math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, assuming the weak Schanuel Conjecture (WSC), we prove that for any collection of pairwise non-arithmetically equivalent totally real number fields, the residues at $s=1$ of their Dedekind zeta functions form a linearly independent set over the field of algebraic numbers. As a corollary, we obtain that, under WSC, two totally real number fields have the same regulator if and only if they have the same class number and Dedekind zeta function. We also prove that, under WSC, the isometry and similarity classes of the log unit lattice of a real Galois number field of degree $[K:\Q]\geq 4$, characterize the isomorphism class of said field. All of our results follow from establishing that, under WSC, any Gram matrix of the log unit lattice of a real Galois number field yields a generic point of certain closed irreducible $\Q$-subvariety of the space of symmetric matrices of appropriate size.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 04:26:27 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 19:24:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cruz",
"José",
""
]
] |
2512.03434
|
Zihao Ren
|
Zihao Ren, Daniel Quevedo, Salah Sukkarieh, and Guodong Shi
|
Quantum Encrypted Control of Networked Systems
|
17 pages
| null | null | null |
quant-ph cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Encrypted control has been extensively studied to ensure the confidentiality of system states and control inputs for networked control systems. This paper presents a computationally efficient encrypted control framework for networked systems enabled by quantum communication. A quantum channel between sensors and actuators is used to generate identical secret keys, whose security is further enhanced through quantum key distribution. These keys enable lightweight encryption and decryption while preserving confidentiality and control accuracy. We develop a novel encryption-decryption architecture for state-feedback control of linear systems based on quantum keys, and characterize the impact of quantum state errors on closed-loop stability. In particular, we establish the existence of a critical threshold on intrinsic quantum noise below which stability is guaranteed. In contrast to classical encrypted control schemes, which may collapse under a single key-bit error, the proposed quantum encrypted control exhibits strong robustness to key imperfections. We further adopt quantization techniques to address the scenarios with limited communication bits in practical situations, and implement privacy protection for quantum keys based on a stochastic quantizer. These results demonstrate that integrating quantum technologies into control systems in a nontrivial and principled manner, even at their current level of maturity, can yield substantial performance gains in reducing computational complexity and improving resilience to key errors while ensuring security against multiple eavesdropping sources.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 04:27:39 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:32:15 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ren",
"Zihao",
""
],
[
"Quevedo",
"Daniel",
""
],
[
"Sukkarieh",
"Salah",
""
],
[
"Shi",
"Guodong",
""
]
] |
2512.03454
|
Haicheng Liao
|
Haicheng Liao, Huanming Shen, Bonan Wang, Yongkang Li, Yihong Tang, Chengyue Wang, Dingyi Zhuang, Kehua Chen, Hai Yang, Chengzhong Xu, Zhenning Li
|
Think Before You Drive: World Model-Inspired Multimodal Grounding for Autonomous Vehicles
| null | null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Interpreting natural-language commands to localize target objects is critical for autonomous driving (AD). Existing visual grounding (VG) methods for autonomous vehicles (AVs) typically struggle with ambiguous, context-dependent instructions, as they lack reasoning over 3D spatial relations and anticipated scene evolution. Grounded in the principles of world models, we propose ThinkDeeper, a framework that reasons about future spatial states before making grounding decisions. At its core is a Spatial-Aware World Model (SA-WM) that learns to reason ahead by distilling the current scene into a command-aware latent state and rolling out a sequence of future latent states, providing forward-looking cues for disambiguation. Complementing this, a hypergraph-guided decoder then hierarchically fuses these states with the multimodal input, capturing higher-order spatial dependencies for robust localization. In addition, we present DrivePilot, a multi-source VG dataset in AD, featuring semantic annotations generated by a Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT)-prompted LLM pipeline. Extensive evaluations on six benchmarks, ThinkDeeper ranks #1 on the Talk2Car leaderboard and surpasses state-of-the-art baselines on DrivePilot, MoCAD, and RefCOCO/+/g benchmarks. Notably, it shows strong robustness and efficiency in challenging scenes (long-text, multi-agent, ambiguity) and retains superior performance even when trained on 50% of the data.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 05:14:16 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 07:03:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liao",
"Haicheng",
""
],
[
"Shen",
"Huanming",
""
],
[
"Wang",
"Bonan",
""
],
[
"Li",
"Yongkang",
""
],
[
"Tang",
"Yihong",
""
],
[
"Wang",
"Chengyue",
""
],
[
"Zhuang",
"Dingyi",
""
],
[
"Chen",
"Kehua",
""
],
[
"Yang",
"Hai",
""
],
[
"Xu",
"Chengzhong",
""
],
[
"Li",
"Zhenning",
""
]
] |
2512.03477
|
Zijian Gu
|
Zijian Gu, Yuxi Liu, Zhenhao Zhang, Song Wang
|
Fairness-Aware Fine-Tuning of Vision-Language Models for Medical Glaucoma Diagnosis
|
10 pages, 3 tables
| null | null | null |
cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Vision-language models achieve expert-level performance on medical imaging tasks but exhibit significant diagnostic accuracy disparities across demographic groups. We introduce fairness-aware Low-Rank Adaptation for medical VLMs, combining parameter efficiency with explicit fairness optimization. Our key algorithmic contribution is a differentiable MaxAccGap loss that enables end-to-end optimization of accuracy parity across demographic groups. We propose three methods: FR-LoRA integrates MaxAccGap regularization into the training objective, GR-LoRA applies inverse frequency weighting to balance gradient contributions, and Hybrid-LoRA combines both mechanisms. Evaluated on 10,000 glaucoma fundus images, GR-LoRA reduces diagnostic accuracy disparities by 69% while maintaining 53.15% overall accuracy. Ablation studies reveal that strong regularization strength achieves optimal fairness with minimal accuracy trade-off, and race-specific optimization yields 60% disparity reduction. Our approach requires only 0.24% trainable parameters, enabling practical deployment of fair medical AI in resource-constrained healthcare settings.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 06:09:14 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 17:17:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gu",
"Zijian",
""
],
[
"Liu",
"Yuxi",
""
],
[
"Zhang",
"Zhenhao",
""
],
[
"Wang",
"Song",
""
]
] |
2512.04128
|
Bhola Dwivedi Prof
|
Klaus Wilhelm, Bhola N. Dwivedi, Karsten Muller
|
Questions related to the Deflection of Light by Gravity determined by Soldner, Einstein and Schwarzschild
|
20 pages; Version 2
| null | null | null |
physics.hist-ph astro-ph.SR gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Before we discuss the deflection of light in a gravitational field, we give a brief overview of some basic physical formulas on photon properties, generation and propagation. The much debated problems of the redshift and the photon propagation in a gravitational field is then considered and applied to the calculation of the speed of light. Many citations are given in direct quotations to avoid any misunderstandings. If the quotations are in German, an English translation is provided. Based on this speed, calculated and measured results are recalled on the deflection of light, with emphasis on the deflection near the Sun. We conclude that the speed of light and the deflection angle can be determined by energy and momentum conservation principles.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 03:33:55 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:16:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wilhelm",
"Klaus",
""
],
[
"Dwivedi",
"Bhola N.",
""
],
[
"Muller",
"Karsten",
""
]
] |
2512.04215
|
Marzia Mazzotta
|
Marzia Mazzotta, Paola Stefanelli, Magdalena Wiertel
|
Quasi racks, quasi bijective and quasi non-degenerate set-theoretic solutions of the Yang-Baxter equation
| null | null | null | null |
math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
This work initiates a systematic study of the class of quasi bijective and quasi non-degenerate solutions to the set-theoretic Yang-Baxter equation. The motivation stems from the observation that solutions that arise from dual weak braces belong to these classes. The notions of quasi rack and derived solution are introduced and examined, extending the classical definitions. Additionally, a family of quasi left non-degenerate solutions is described in terms of quasi racks and g-twists, analogous to the left non-degenerate case. Furthermore, we completely characterize a class of quasi racks that are Plonka sum of racks.
|
[
{
"version": "v1",
"created": "Wed, 03 Dec 2025 19:39:47 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 15:18:03 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mazzotta",
"Marzia",
""
],
[
"Stefanelli",
"Paola",
""
],
[
"Wiertel",
"Magdalena",
""
]
] |
2512.04432
|
Ye Tian
|
Ye Tian, Souvik Maity, Jingyu Li, Yuancai Wu, Shan Sha, Yutie Liang, Aiqiang Guo, Yuxiang Zhao, and Dexu Lin
|
Design and Performance Simulation of the Electromagnetic Calorimeter at EicC
| null | null | null | null |
physics.ins-det hep-ex
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The electromagnetic calorimeter (ECAL) is a key detector component for precise electron and photon measurements in electron-ion collision experiments. At the Electron-Ion Collider in China (EicC), high-performance calorimetry is essential for exploring the internal structure of nucleons and studying the dynamics of quarks and gluons within quantum chromodynamics (QCD). This paper presents the optimized design and performance simulation of the EicC ECAL system. The ECAL consists of three specialized sections tailored to distinct detection environments: (1) an electron-Endcap employing high-resolution pure Cesium Iodide (pCsI) crystals, (2) a central barrel, and (3) an ion-Endcap, both adopting a cost-effective Shashlik-style sampling calorimeter with improved light yield. Each segment's geometry and material composition have been systematically optimized through Geant4 simulations to achieve excellent energy and position resolutions as well as strong electron-pion discrimination. The simulated performance indicates that the ECAL can achieve energy resolutions of 2 percent divided by sqrt(E) for pCsI crystals and 5 percent divided by sqrt(E) for Shashlik modules, meeting the design goals of the EicC detector.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 03:55:34 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 01:41:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tian",
"Ye",
""
],
[
"Maity",
"Souvik",
""
],
[
"Li",
"Jingyu",
""
],
[
"Wu",
"Yuancai",
""
],
[
"Sha",
"Shan",
""
],
[
"Liang",
"Yutie",
""
],
[
"Guo",
"Aiqiang",
""
],
[
"Zhao",
"Yuxiang",
""
],
[
"Lin",
"Dexu",
""
]
] |
2512.04629
|
Siqi Fan
|
Chenyang Zuo, Siqi Fan, Zaiqing Nie
|
BioMedGPT-Mol: Multi-task Learning for Molecular Understanding and Generation
| null | null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Molecules play a crucial role in biomedical research and discovery, particularly in the field of small molecule drug development. Given the rapid advancements in large language models, especially the recent emergence of reasoning models, it is natural to explore how a general-purpose language model can be efficiently adapted for molecular science applications. In this work, we introduce BioMedGPT-Mol, a molecular language model designed to support molecular understanding and generation tasks. By curating and unifying existing public instruction datasets, we have assembled a large-scale, comprehensive, and high-quality training dataset. The model is then fine-tuned through a meticulously designed multi-task learning framework. On a consolidated benchmark derived from LlaSMol, TOMG-Bench, and MuMOInstruct, BioMedGPT-Mol achieves remarkable performance. Our experimental results demonstrate that a general-purpose reasoning model can be effectively and efficiently post-trained into a professional molecular language model through a well-structured multi-task curriculum. Leveraging these capabilities, we further apply the model to multi-step retrosynthetic planning, achieving state-of-the-art performance on RetroBench and demonstrating its superior efficacy as an end-to-end retrosynthetic planner. We anticipate that our approach can be extended to other biomedical scientific domains.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 10:00:16 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:24:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zuo",
"Chenyang",
""
],
[
"Fan",
"Siqi",
""
],
[
"Nie",
"Zaiqing",
""
]
] |
2512.04721
|
Diego A. Souza
|
Felipe W. Chaves-Silva, Marcos G. Ferreira-Silva and Diego A. Souza
|
Optimal cost for the null controllability of the Stokes system with controls having $n-1$ components and applications
|
25 pages. Comments are welcome
| null | null | null |
math.AP math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we investigate the optimal cost of null controllability for the $n$-dimensional Stokes system when the control acts on $n-1$ scalar components. We establish a novel spectral estimate for low frequencies of the Stokes operator, involving solely $n-1$ components, and use it to show that the cost of controllability with controls having $n-1$ components remains of the same order in time as in the case of controls with $n$ components, namely $O(e^{C/T})$, i.e. the cost of null controllability is not affected by the absence of one component of the control. We also give several applications of our results.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 12:05:52 GMT"
},
{
"version": "v2",
"created": "Fri, 05 Dec 2025 05:23:50 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 21:50:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chaves-Silva",
"Felipe W.",
""
],
[
"Ferreira-Silva",
"Marcos G.",
""
],
[
"Souza",
"Diego A.",
""
]
] |
2512.04906
|
Liming Dou
|
Jiahua Wu, Huifang Xie, Liming Dou, Yanli Ai, Tinggui Wang, Xinwen Shu, Ning Jiang, Luis C. Ho and Junhui Fan
|
Long-term Mid-infrared Color Variations of Narrow-Line Seyfert 1 Galaxies
|
Accepted for publication in The Astrophysical Journal
| null |
10.3847/1538-4357/ae28c8
| null |
astro-ph.GA astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a systematic investigation of long-term mid-infrared (MIR) color variability in 1,718 Narrow-Line Seyfert 1 galaxies (NLSy1s) using 14-year \textit{WISE}/NEOWISE monitoring data. Through Pearson correlation analysis between photometric magnitude and color, we identify: (1) a radio-quiet NLSy1 (RQ-NLSy1) population comprising 230 bluer-when-brighter (BWB) sources, 131 redder-when-brighter (RWB) sources, and 1,323 objects showing weak or statistically insignificant color variations; and (2) a radio-loud NLSy1 (RL-NLSy1) population containing 5 BWBs, 2 RWBs, and 27 sources with weak/no color variations. Our analysis reveals that the BWB tendency strengthens significantly in galaxies with redder mean MIR colors $\rm \left<W1-W2\right>$ and lower starlight contamination. Furthermore, this color-change pattern demonstrates that the most bolometric luminous sources exhibit the most pronounced BWB behavior. While similar trends exist for black hole mass and Eddington ratio, bolometric luminosity appears to be the primary physical driver. Potential origins of these variations (e.g., host galaxy contribution, accretion disk variability, and dust reprocessing) are discussed. We conclude that temperature-dependent dust reprocessing dominates the observed BWB, RWB, and no/weak variation patterns. This interpretation may also apply to similar MIR color variations observed in other extragalactic MIR transients, such as tidal disruption events, ambiguous nuclear transients, and changing-look AGNs. In addition, we find no significant difference in long-term MIR color variations between RL-NLSy1s and RQ-NLSy1s, however, RL-NLSy1s show significantly greater dispersion in intrinsic variability amplitude compared to RQ-NLSy1s due to jet-induced complexity, where non-thermal synchrotron emission from relativistic jets obscures thermal dust signatures.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 15:36:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wu",
"Jiahua",
""
],
[
"Xie",
"Huifang",
""
],
[
"Dou",
"Liming",
""
],
[
"Ai",
"Yanli",
""
],
[
"Wang",
"Tinggui",
""
],
[
"Shu",
"Xinwen",
""
],
[
"Jiang",
"Ning",
""
],
[
"Ho",
"Luis C.",
""
],
[
"Fan",
"Junhui",
""
]
] |
2512.04919
|
D\'aniel Virosztek
|
Gergely Bunth, J\'ozsef Pitrik, Tam\'as Titkos, D\'aniel Virosztek
|
The swap transpose on couplings translates to Petz' recovery map on quantum channels
|
11 pages. v2: References to the closely related works [Duvenhage et al., Quantum 9, 1743 (2025)] and [Parzygnat et al., PRX Quantum 4, 020334 (2023)] have been included
| null | null | null |
math-ph math.MP math.OA quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In [Ann. Henri Poincar\'e, {\bf 22} (2021), 3199-3234], De Palma and Trevisan described a one-to-one correspondence between quantum couplings and quantum channels realizing transport between states. The aim of this short note is to demonstrate that taking the Petz recovery map for a given channel and initial state is precisely the counterpart of the swap transpose operation on couplings. That is, the swap transpose of the coupling $\Pi_{\Phi}$ corresponding to the channel $\Phi$ and initial state $\rho$ is the coupling $\Pi_{rec}$ corresponding to the Petz recovery map $\Phi_{rec}.$
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 15:47:32 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:49:54 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bunth",
"Gergely",
""
],
[
"Pitrik",
"József",
""
],
[
"Titkos",
"Tamás",
""
],
[
"Virosztek",
"Dániel",
""
]
] |
2512.04948
|
Zihan Wang
|
Arkady A. Tseytlin and Zihan Wang
|
On world-volume supersymmetry of supermembrane action in static gauge
|
20 pages, v2: references added
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We review and elaborate on the issue of 3d world volume supersymmetry that appears as a residual part of global target space supersymmetry in the BST supermembrane action. While there is no direct ``spinning membrane'' analog of the world-volume supersymmetric spinning string action that could be obtained by coupling $D$ copies of 3d scalar multiplet to 3d supergravity, we discuss how one may construct an $N=1$ 3d supersymmetric analog of the derivative expansion of the bosonic membrane action in static gauge. We compare the resulting $N=1$ supersymmetric action for eight 3d scalar multiplets to the $N=8$ 3d supersymmetric action describing the $D=11$ supermembrane in the static gauge. The two actions are not equivalent which is related to the fact that the full $N=8$ supersymmetry of the static-gauge $D=11$ supermembrane action can be realised only if the fermions are described by an $SO(8)$ spinor rather than vector. The two actions are still directly related in special dimensions $D=4$ and 5. We also compute the one-loop world-volume scattering amplitudes for the two theories, finding that they indeed agree for $D=4,5$ but disagree for $D=11$.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 16:14:57 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 20:32:49 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tseytlin",
"Arkady A.",
""
],
[
"Wang",
"Zihan",
""
]
] |
2512.04993
|
Yisai Xue
|
Zhuo Wu, Yisai Xue
|
Edge density and minimum degree thresholds for $H$-free graphs with unbounded chromatic number
|
13 pages, 2 figures
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The chromatic threshold $\delta_\chi(H)$ of a graph $H$ is the infimum of $d>0$ such that the chromatic number of every $n$-vertex $H$-free graph with minimum degree at least $dn$ is bounded in terms of $H$ and $d$. A breakthrough result of Allen, B\"ottcher, Griffiths, Kohayakawa, and Morris determined $\delta_\chi(H)$ for every graph $H$; in particular, if $\chi(H)=r\ge 3$, then $\delta_\chi(H) \in\{\frac{r-3}{r-2},~\frac{2 r-5}{2 r-3},~\frac{r-2}{r-1}\}$.
In this paper we investigate the trade-off between minimum degree and edge density in the critical window around the chromatic threshold. For a fixed graph $H$ with $\chi(H)=r$, allowing a constant deficit below $\delta_\chi(H)$, we prove sharp (up to lower-order terms) upper bounds on the edge density of $n$-vertex $H$-free graphs whose chromatic number diverges. Equivalently, within this degree regime we show that a suitable global bound on the number of edges forces the chromatic number to remain bounded. Our results thus quantify how global edge density can compensate for a deficit in the local minimum-degree condition near $\delta_\chi(H)$; more specifically, we obtain explicit bounds in two of the three possible cases arising in the trichotomy of $\delta_\chi(H)$. Our extremal constructions -- based on Erd\H{o}s graphs and blowups of Borsuk--Hajnal graphs -- show that these bounds are best possible up to $o(n^2)$ terms.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 17:06:09 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 12:48:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wu",
"Zhuo",
""
],
[
"Xue",
"Yisai",
""
]
] |
2512.05001
|
Vincent Pilaud
|
Vincent Pilaud
|
Plumbing bijections
|
5 pages, 5 figures. Version 2: more references added
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The legendary Mario and Luigi show us that whether you slap in the crossings as early as a warp pipe can shoot you or as late as the very last bend, the water system in Yoshi Hill comes out exactly the same!
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 17:12:55 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:50:35 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Pilaud",
"Vincent",
""
]
] |
2512.05083
|
Sebouh Paul
|
S.J. Paul, M. Arratia, H. Hakobyan, W. Brooks, A. Acar, P. Achenbach, J.S. Alvarado, W.R. Armstrong, N.A. Baltzell, L. Barion, M. Bashkanov, M. Battaglieri, F. Benmokhtar, A. Bianconi, A.S. Biselli, F. Boss\`u, S. Boiarinov, K.-T. Brinkmann, W.J. Briscoe, V. Burkert, T. Cao, D.S. Carman, P. Chatagnon, H. Chinchay, G. Ciullo, P.L. Cole, A. D'Angelo, N. Dashyan, R. De Vita, A. Deur, S. Diehl, C. Djalali, R. Dupre, H. Egiyan, A. El Alaoui, L. Elouadrhiri, P. Eugenio, M. Farooq, S. Fegan, A. Filippi, C. Fogler, G. Gavalian, G.P. Gilfoyle, R.W. Gothe, B. Gualtieri, M. Hattawy, F. Hauenstein, T.B. Hayward, M. Hoballah, M. Holtrop, Yu-Chun Hung, Y. Ilieva, D.G. Ireland, E.L. Isupov, D. Jenkins, H.S. Jo, D. Keller, M. Khandaker, A. Kim, V. Klimenko, I. Korover, A. Kripko, V. Kubarovsky, L. Lanza, S. Lee, P. Lenisa, X. Li, D. Marchand, V. Mascagna, B. McKinnon, T. Mineeva, V. Mokeev, E.F. Molina Cardenas, C. Munoz Camacho, P. Nadel-Turonski, T. Nagorna, K. Neupane, S. Niccolai, G. Niculescu, M. Osipenko, A.I. Ostrovidov, M. Ouillon, P. Pandey, M. Paolone, L.L. Pappalardo, R. Paremuzyan, E. Pasyuk, C. Paudel, W. Phelps, N. Pilleux, P.S.H. Vaishnavi, S. Polcher Rafael, L. Polizzi, J.W. Price, Y. Prok, A. Radic, T. Reed, J. Richards, M. Ripani, J. Ritman, G. Rosner, S. Schadmand, A. Schmidt, R.A. Schumacher, Y. Sharabian, S. Shrestha, E. Sidoretti, D. Sokhan, N. Sparveris, M. Spreafico, S. Stepanyan, I.I. Strakovsky, S. Strauch, M. Tenorio, F. Touchte Codjo, R. Tyson, M. Ungaro, S. Vallarino, C. Velasquez, L. Venturelli, H. Voskanyan, E. Voutier, Y. Wang, D.P. Watts, U. Weerasinghe, X. Wei, M.H. Wood, L. Xu, Z. Xu, M. Zurek (the CLAS Collaboration)
|
First Study of the Nuclear Response to Fast Hadrons via Angular Correlations between Pions and Slow Protons in Electron-Nucleus Scattering
| null | null | null | null |
nucl-ex
|
http://creativecommons.org/licenses/by/4.0/
|
We report on the first measurement of angular correlations between high-energy pions and slow protons in electron-nucleus ($eA$) scattering, providing a new probe of how a nucleus responds to a fast-moving quark. The experiment employed the CLAS detector with a 5-GeV electron beam incident on deuterium, carbon, iron, and lead targets. For heavier nuclei, the pion-proton correlation function is more spread-out in azimuth than for lighter ones, and this effect is more pronounced in the $\pi p$ channel than in earlier $\pi\pi$ studies. The proton-to-pion yield ratio likewise rises with nuclear mass, although the increase appears to saturate for the heaviest targets. These trends are qualitatively reproduced by state-of-the-art $eA$ event generators, including BeAGLE, eHIJING, and GiBUU, indicating that current descriptions of target fragmentation rest on sound theoretical footing. At the same time, the precision of our data exposes model-dependent discrepancies, delineating a clear path for future improvements in the treatment of cold-nuclear matter effects in $eA$ scattering.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 18:49:57 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:28:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Paul",
"S. J.",
"",
"the CLAS Collaboration"
],
[
"Arratia",
"M.",
"",
"the CLAS Collaboration"
],
[
"Hakobyan",
"H.",
"",
"the CLAS Collaboration"
],
[
"Brooks",
"W.",
"",
"the CLAS Collaboration"
],
[
"Acar",
"A.",
"",
"the CLAS Collaboration"
],
[
"Achenbach",
"P.",
"",
"the CLAS Collaboration"
],
[
"Alvarado",
"J. S.",
"",
"the CLAS Collaboration"
],
[
"Armstrong",
"W. R.",
"",
"the CLAS Collaboration"
],
[
"Baltzell",
"N. A.",
"",
"the CLAS Collaboration"
],
[
"Barion",
"L.",
"",
"the CLAS Collaboration"
],
[
"Bashkanov",
"M.",
"",
"the CLAS Collaboration"
],
[
"Battaglieri",
"M.",
"",
"the CLAS Collaboration"
],
[
"Benmokhtar",
"F.",
"",
"the CLAS Collaboration"
],
[
"Bianconi",
"A.",
"",
"the CLAS Collaboration"
],
[
"Biselli",
"A. S.",
"",
"the CLAS Collaboration"
],
[
"Bossù",
"F.",
"",
"the CLAS Collaboration"
],
[
"Boiarinov",
"S.",
"",
"the CLAS Collaboration"
],
[
"Brinkmann",
"K. -T.",
"",
"the CLAS Collaboration"
],
[
"Briscoe",
"W. J.",
"",
"the CLAS Collaboration"
],
[
"Burkert",
"V.",
"",
"the CLAS Collaboration"
],
[
"Cao",
"T.",
"",
"the CLAS Collaboration"
],
[
"Carman",
"D. S.",
"",
"the CLAS Collaboration"
],
[
"Chatagnon",
"P.",
"",
"the CLAS Collaboration"
],
[
"Chinchay",
"H.",
"",
"the CLAS Collaboration"
],
[
"Ciullo",
"G.",
"",
"the CLAS Collaboration"
],
[
"Cole",
"P. L.",
"",
"the CLAS Collaboration"
],
[
"D'Angelo",
"A.",
"",
"the CLAS Collaboration"
],
[
"Dashyan",
"N.",
"",
"the CLAS Collaboration"
],
[
"De Vita",
"R.",
"",
"the CLAS Collaboration"
],
[
"Deur",
"A.",
"",
"the CLAS Collaboration"
],
[
"Diehl",
"S.",
"",
"the CLAS Collaboration"
],
[
"Djalali",
"C.",
"",
"the CLAS Collaboration"
],
[
"Dupre",
"R.",
"",
"the CLAS Collaboration"
],
[
"Egiyan",
"H.",
"",
"the CLAS Collaboration"
],
[
"Alaoui",
"A. El",
"",
"the CLAS Collaboration"
],
[
"Elouadrhiri",
"L.",
"",
"the CLAS Collaboration"
],
[
"Eugenio",
"P.",
"",
"the CLAS Collaboration"
],
[
"Farooq",
"M.",
"",
"the CLAS Collaboration"
],
[
"Fegan",
"S.",
"",
"the CLAS Collaboration"
],
[
"Filippi",
"A.",
"",
"the CLAS Collaboration"
],
[
"Fogler",
"C.",
"",
"the CLAS Collaboration"
],
[
"Gavalian",
"G.",
"",
"the CLAS Collaboration"
],
[
"Gilfoyle",
"G. P.",
"",
"the CLAS Collaboration"
],
[
"Gothe",
"R. W.",
"",
"the CLAS Collaboration"
],
[
"Gualtieri",
"B.",
"",
"the CLAS Collaboration"
],
[
"Hattawy",
"M.",
"",
"the CLAS Collaboration"
],
[
"Hauenstein",
"F.",
"",
"the CLAS Collaboration"
],
[
"Hayward",
"T. B.",
"",
"the CLAS Collaboration"
],
[
"Hoballah",
"M.",
"",
"the CLAS Collaboration"
],
[
"Holtrop",
"M.",
"",
"the CLAS Collaboration"
],
[
"Hung",
"Yu-Chun",
"",
"the CLAS Collaboration"
],
[
"Ilieva",
"Y.",
"",
"the CLAS Collaboration"
],
[
"Ireland",
"D. G.",
"",
"the CLAS Collaboration"
],
[
"Isupov",
"E. L.",
"",
"the CLAS Collaboration"
],
[
"Jenkins",
"D.",
"",
"the CLAS Collaboration"
],
[
"Jo",
"H. S.",
"",
"the CLAS Collaboration"
],
[
"Keller",
"D.",
"",
"the CLAS Collaboration"
],
[
"Khandaker",
"M.",
"",
"the CLAS Collaboration"
],
[
"Kim",
"A.",
"",
"the CLAS Collaboration"
],
[
"Klimenko",
"V.",
"",
"the CLAS Collaboration"
],
[
"Korover",
"I.",
"",
"the CLAS Collaboration"
],
[
"Kripko",
"A.",
"",
"the CLAS Collaboration"
],
[
"Kubarovsky",
"V.",
"",
"the CLAS Collaboration"
],
[
"Lanza",
"L.",
"",
"the CLAS Collaboration"
],
[
"Lee",
"S.",
"",
"the CLAS Collaboration"
],
[
"Lenisa",
"P.",
"",
"the CLAS Collaboration"
],
[
"Li",
"X.",
"",
"the CLAS Collaboration"
],
[
"Marchand",
"D.",
"",
"the CLAS Collaboration"
],
[
"Mascagna",
"V.",
"",
"the CLAS Collaboration"
],
[
"McKinnon",
"B.",
"",
"the CLAS Collaboration"
],
[
"Mineeva",
"T.",
"",
"the CLAS Collaboration"
],
[
"Mokeev",
"V.",
"",
"the CLAS Collaboration"
],
[
"Cardenas",
"E. F. Molina",
"",
"the CLAS Collaboration"
],
[
"Camacho",
"C. Munoz",
"",
"the CLAS Collaboration"
],
[
"Nadel-Turonski",
"P.",
"",
"the CLAS Collaboration"
],
[
"Nagorna",
"T.",
"",
"the CLAS Collaboration"
],
[
"Neupane",
"K.",
"",
"the CLAS Collaboration"
],
[
"Niccolai",
"S.",
"",
"the CLAS Collaboration"
],
[
"Niculescu",
"G.",
"",
"the CLAS Collaboration"
],
[
"Osipenko",
"M.",
"",
"the CLAS Collaboration"
],
[
"Ostrovidov",
"A. I.",
"",
"the CLAS Collaboration"
],
[
"Ouillon",
"M.",
"",
"the CLAS Collaboration"
],
[
"Pandey",
"P.",
"",
"the CLAS Collaboration"
],
[
"Paolone",
"M.",
"",
"the CLAS Collaboration"
],
[
"Pappalardo",
"L. L.",
"",
"the CLAS Collaboration"
],
[
"Paremuzyan",
"R.",
"",
"the CLAS Collaboration"
],
[
"Pasyuk",
"E.",
"",
"the CLAS Collaboration"
],
[
"Paudel",
"C.",
"",
"the CLAS Collaboration"
],
[
"Phelps",
"W.",
"",
"the CLAS Collaboration"
],
[
"Pilleux",
"N.",
"",
"the CLAS Collaboration"
],
[
"Vaishnavi",
"P. S. H.",
"",
"the CLAS Collaboration"
],
[
"Rafael",
"S. Polcher",
"",
"the CLAS Collaboration"
],
[
"Polizzi",
"L.",
"",
"the CLAS Collaboration"
],
[
"Price",
"J. W.",
"",
"the CLAS Collaboration"
],
[
"Prok",
"Y.",
"",
"the CLAS Collaboration"
],
[
"Radic",
"A.",
"",
"the CLAS Collaboration"
],
[
"Reed",
"T.",
"",
"the CLAS Collaboration"
],
[
"Richards",
"J.",
"",
"the CLAS Collaboration"
],
[
"Ripani",
"M.",
"",
"the CLAS Collaboration"
],
[
"Ritman",
"J.",
"",
"the CLAS Collaboration"
],
[
"Rosner",
"G.",
"",
"the CLAS Collaboration"
],
[
"Schadmand",
"S.",
"",
"the CLAS Collaboration"
],
[
"Schmidt",
"A.",
"",
"the CLAS Collaboration"
],
[
"Schumacher",
"R. A.",
"",
"the CLAS Collaboration"
],
[
"Sharabian",
"Y.",
"",
"the CLAS Collaboration"
],
[
"Shrestha",
"S.",
"",
"the CLAS Collaboration"
],
[
"Sidoretti",
"E.",
"",
"the CLAS Collaboration"
],
[
"Sokhan",
"D.",
"",
"the CLAS Collaboration"
],
[
"Sparveris",
"N.",
"",
"the CLAS Collaboration"
],
[
"Spreafico",
"M.",
"",
"the CLAS Collaboration"
],
[
"Stepanyan",
"S.",
"",
"the CLAS Collaboration"
],
[
"Strakovsky",
"I. I.",
"",
"the CLAS Collaboration"
],
[
"Strauch",
"S.",
"",
"the CLAS Collaboration"
],
[
"Tenorio",
"M.",
"",
"the CLAS Collaboration"
],
[
"Codjo",
"F. Touchte",
"",
"the CLAS Collaboration"
],
[
"Tyson",
"R.",
"",
"the CLAS Collaboration"
],
[
"Ungaro",
"M.",
"",
"the CLAS Collaboration"
],
[
"Vallarino",
"S.",
"",
"the CLAS Collaboration"
],
[
"Velasquez",
"C.",
"",
"the CLAS Collaboration"
],
[
"Venturelli",
"L.",
"",
"the CLAS Collaboration"
],
[
"Voskanyan",
"H.",
"",
"the CLAS Collaboration"
],
[
"Voutier",
"E.",
"",
"the CLAS Collaboration"
],
[
"Wang",
"Y.",
"",
"the CLAS Collaboration"
],
[
"Watts",
"D. P.",
"",
"the CLAS Collaboration"
],
[
"Weerasinghe",
"U.",
"",
"the CLAS Collaboration"
],
[
"Wei",
"X.",
"",
"the CLAS Collaboration"
],
[
"Wood",
"M. H.",
"",
"the CLAS Collaboration"
],
[
"Xu",
"L.",
"",
"the CLAS Collaboration"
],
[
"Xu",
"Z.",
"",
"the CLAS Collaboration"
],
[
"Zurek",
"M.",
"",
"the CLAS Collaboration"
]
] |
2512.05094
|
James Ni
|
James Ni, Zekai Wang, Wei Lin, Amir Bar, Yann LeCun, Trevor Darrell, Jitendra Malik, Roei Herzig
|
From Generated Human Videos to Physically Plausible Robot Trajectories
|
For project website, see https://genmimic.github.io
| null | null | null |
cs.RO cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Video generation models are rapidly improving in their ability to synthesize human actions in novel contexts, holding the potential to serve as high-level planners for contextual robot control. To realize this potential, a key research question remains open: how can a humanoid execute the human actions from generated videos in a zero-shot manner? This challenge arises because generated videos are often noisy and exhibit morphological distortions that make direct imitation difficult compared to real video. To address this, we introduce a two-stage pipeline. First, we lift video pixels into a 4D human representation and then retarget to the humanoid morphology. Second, we propose GenMimic-a physics-aware reinforcement learning policy conditioned on 3D keypoints, and trained with symmetry regularization and keypoint-weighted tracking rewards. As a result, GenMimic can mimic human actions from noisy, generated videos. We curate GenMimicBench, a synthetic human-motion dataset generated using two video generation models across a spectrum of actions and contexts, establishing a benchmark for assessing zero-shot generalization and policy robustness. Extensive experiments demonstrate improvements over strong baselines in simulation and confirm coherent, physically stable motion tracking on a Unitree G1 humanoid robot without fine-tuning. This work offers a promising path to realizing the potential of video generation models as high-level policies for robot control.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 18:56:03 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 17:37:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ni",
"James",
""
],
[
"Wang",
"Zekai",
""
],
[
"Lin",
"Wei",
""
],
[
"Bar",
"Amir",
""
],
[
"LeCun",
"Yann",
""
],
[
"Darrell",
"Trevor",
""
],
[
"Malik",
"Jitendra",
""
],
[
"Herzig",
"Roei",
""
]
] |
2512.05104
|
Jiaqi Ma
|
Jiaqi Ma, Shengkai Hu, Xu Zhang, Jun Wan, Jiaxing Huang, Lefei Zhang, Salman Khan
|
EvoIR: Towards All-in-One Image Restoration via Evolutionary Frequency Modulation
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
All-in-One Image Restoration (AiOIR) tasks often involve diverse degradation that require robust and versatile strategies. However, most existing approaches typically lack explicit frequency modeling and rely on fixed or heuristic optimization schedules, which limit the generalization across heterogeneous degradation. To address these limitations, we propose EvoIR, an AiOIR-specific framework that introduces evolutionary frequency modulation for dynamic and adaptive image restoration. Specifically, EvoIR employs the Frequency-Modulated Module (FMM) that decomposes features into high- and low-frequency branches in an explicit manner and adaptively modulates them to enhance both structural fidelity and fine-grained details. Central to EvoIR, an Evolutionary Optimization Strategy (EOS) iteratively adjusts frequency-aware objectives through a population-based evolutionary process, dynamically balancing structural accuracy and perceptual fidelity. Its evolutionary guidance further mitigates gradient conflicts across degradation and accelerates convergence. By synergizing FMM and EOS, EvoIR yields greater improvements than using either component alone, underscoring their complementary roles. Extensive experiments on multiple benchmarks demonstrate that EvoIR outperforms state-of-the-art AiOIR methods.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 18:59:10 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:59:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ma",
"Jiaqi",
""
],
[
"Hu",
"Shengkai",
""
],
[
"Zhang",
"Xu",
""
],
[
"Wan",
"Jun",
""
],
[
"Huang",
"Jiaxing",
""
],
[
"Zhang",
"Lefei",
""
],
[
"Khan",
"Salman",
""
]
] |
2512.05142
|
Anil Yadav dr
|
Anil Kumar Yadav, S. H. Shekh, N. Myrzakulov
|
Effective $\Lambda$CDM model emerging from $f(Q,T)$ under a special EOS limit in symmetric cosmology with Bayesian and ANN observational constraints
|
14 Pages, 10 Figures
| null | null | null |
gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we investigate the cosmological consequences of an effective $f(Q)$ model emerging from the more general $f(Q,T)$ gravity theory under the special equation-of-state condition $\rho + p = 0$. Under this limit, the field equations yield the constraint $F(Q,T)H(t)=C$, implying that the function $F=f_Q$ becomes purely dependent on the nonmetricity scalar $Q$, and the background evolution mimics that of the standard $\Lambda$CDM model. We derive the resulting functional forms of $f(Q)$, obtain the corresponding effective cosmological constant, and analyze the physical nature of this reduction. To test the model against observations, we constrain the parameters $H_0$, $\Omega_m$, and $S_8$ using cosmic chronometers (CC), baryon acoustic oscillations (BAO), and Pantheon+ SN Ia datasets. A comparative analysis is performed using both the conventional Bayesian Markov Chain Monte Carlo (MCMC) sampling and a machine-learning based Artificial Neural Network (ANN) emulator. We find that the ANN approach yields tighter posterior constraints while significantly reducing computational time. The model successfully reproduces the observational trends of each dataset and offers insights into the persistent $H_0$ and $S_8$ tensions. Our results indicate that effective nonmetricity-based dark energy scenarios derived from $f(Q,T)$ gravity provide a viable and observationally consistent alternative to $\Lambda$CDM, with future high-precision surveys expected to further distinguish between these frameworks.
|
[
{
"version": "v1",
"created": "Tue, 02 Dec 2025 06:10:46 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 11:27:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yadav",
"Anil Kumar",
""
],
[
"Shekh",
"S. H.",
""
],
[
"Myrzakulov",
"N.",
""
]
] |
2512.05391
|
Qingqiao Hu
|
Qingqiao Hu, Weimin Lyu, Meilong Xu, Kehan Qi, Xiaoling Hu, Saumya Gupta, Jiawei Zhou, Chao Chen
|
LoC-Path: Learning to Compress for Pathology Multimodal Large Language Models
|
20 pages
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Whole Slide Image (WSI) understanding is fundamentally challenging due to its gigapixel scale and the extreme sparsity of diagnostically relevant regions. Unlike human experts who primarily rely on key areas to arrive at a diagnosis, existing slide-level multimodal large language models (MLLMs) for pathology rely on heavy slide-level encoders that process thousands of patch features in a brute-force manner, resulting in excessive computational cost. In this work, we revisit the WSI-language modeling paradigm and show that tile-level features exhibit strong global and local redundancy, whereas only a small subset of tiles are truly task-relevant. Motivated by this observation, we introduce an efficient MLLM framework, called LoC-Path, that replaces the expensive slide-level encoder with redundancy-reducing modules. We first design a Sparse Token Merger (STM) and an MAE-pretrained resampler to remove local redundancy and compress globally redundant tile tokens into a compact slide-level representation set. We then propose a Cross-Attention Routing Adapter (CARA) and a Token Importance Scorer (TIS) to integrate the compressed visual representation with the language model in a computation-efficient manner. Extensive experiments demonstrate that our approach achieves performance comparable to existing state-of-the-art whole-slide MLLMs, while requiring significantly lower computation and memory.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 03:16:46 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:04:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hu",
"Qingqiao",
""
],
[
"Lyu",
"Weimin",
""
],
[
"Xu",
"Meilong",
""
],
[
"Qi",
"Kehan",
""
],
[
"Hu",
"Xiaoling",
""
],
[
"Gupta",
"Saumya",
""
],
[
"Zhou",
"Jiawei",
""
],
[
"Chen",
"Chao",
""
]
] |
2512.05393
|
Ali Tozar
|
Ali Tozar
|
Interaction-Induced Breakdown of Anderson Localization: Thermodynamic Segregation disguised as the Skin Effect
|
12 pages, 4 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the interplay between strong disorder and repulsive interactions in the one-dimensional Fermi-Hubbard model under open boundary conditions. While uncorrelated disorder is widely accepted to localize all single-particle eigenstates, a phenomenon typically reinforced by interactions in the Many-Body Localization (MBL) regime, we report a counter-intuitive breakdown of this paradigm. We demonstrate that strong repulsive interactions can overcome disorder-induced localization, driving the system into a macroscopically segregated phase where spin species accumulate at opposite boundaries. Although this boundary accumulation phenomenologically mimics the Non-Hermitian Skin Effect (NHSE) observed in non-reciprocal systems, our comprehensive analysis reveals a fundamentally different origin. By performing a rigorous control experiment in the Hermitian limit, we prove that the segregation persists without non-reciprocity, identifying many-body energy minimization as the primary driver. This "interaction-induced segregation" manifests as a sharp thermodynamic crossover, characterized by a divergent energy susceptibility, challenging the conventional understanding of disorder-interaction competition in open quantum systems.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 03:18:35 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:20:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tozar",
"Ali",
""
]
] |
2512.05414
|
Akesh Samuditha
|
Akesh Gunathilake, Nadil Karunarathna, Tharusha Bandaranayake, Nisansa de Silva, Surangika Ranathunga
|
LMSpell: Neural Spell Checking for Low-Resource Languages
| null | null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Spell correction is still a challenging problem for low-resource languages (LRLs). While pretrained language models (PLMs) have been employed for spell correction, their use is still limited to a handful of languages, and there has been no proper comparison across PLMs. We present the first empirical study on the effectiveness of PLMs for spell correction, which includes LRLs. We find that Large Language Models (LLMs) outperform their counterparts (encoder-based and encoder-decoder) when the fine-tuning dataset is large. This observation holds even in languages for which the LLM is not pre-trained. We release LMSpell, an easy- to use spell correction toolkit across PLMs. It includes an evaluation function that compensates for the hallucination of LLMs. Further, we present a case study with Sinhala to shed light on the plight of spell correction for LRLs.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 04:14:09 GMT"
},
{
"version": "v2",
"created": "Mon, 08 Dec 2025 02:01:26 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 14:22:42 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gunathilake",
"Akesh",
""
],
[
"Karunarathna",
"Nadil",
""
],
[
"Bandaranayake",
"Tharusha",
""
],
[
"de Silva",
"Nisansa",
""
],
[
"Ranathunga",
"Surangika",
""
]
] |
2512.05465
|
Moupriya Das
|
Samudro Ghosh and Moupriya Das
|
Dynamic hysteresis and transitions controlled by asymmetry in potential barrier shaping
|
8 pages, 4 figures
| null | null | null |
cond-mat.stat-mech
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Our study unveils the precise role of the underlying potential in regulating the fundamental processes of dynamic hysteresis, which manifests in numerous natural and designed systems. We identify that it is possible to induce symmetry breaking in dynamic hysteresis, and consequently to observe dynamic transitions under moderate conditions, which is absent for the symmetric case, if appropriate asymmetry is implemented in the design of the underlying potential. This kind of asymmetry appears through the disparate widths of the two wells of the intrinsic bistable potential governing the dynamics and the barrier separating them. It is characteristically distinct from the potential in which the two minima are energetically dissimilar. Our understanding suggests that only the intrinsic asymmetry of the former type can substantially influence the elemental dynamics of the processes to generate significant effects on the outcomes. Our study presents a novel approach to quantitatively regulate the outputs, to increase or decrease the extent of dynamic hysteresis, based on the requirements, by effectively controlling the proper asymmetry of the intrinsic potential.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 06:46:49 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 13:26:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ghosh",
"Samudro",
""
],
[
"Das",
"Moupriya",
""
]
] |
2512.05514
|
Jarne Moens
|
Jarne Moens, Konstantin Zarembo
|
Wilson loops on the Coulomb branch of $N=4$ super-Yang-Mills
|
24 pages,6 figures
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We study Wilson loops on the Coulomb branch of $N = 4$ super-Yang-Mills theory, by solving for minimal surfaces that connect the contour on the boundary with the D3-brane in the bulk of AdS$_5 \times S^5$. The circular loop undergoes the Gross-Ooguri transition as a function of the radius and angular separation, and we fully map its phase diagram. As a byproduct we find evidence that the expectation value of the straight line is tree-level exact.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 08:18:26 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:32:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Moens",
"Jarne",
""
],
[
"Zarembo",
"Konstantin",
""
]
] |
2512.05543
|
Le Chang
|
Xuezhi Li, Jiancong He, Ming Xie, Xuyang Chen, Le Chang, Li Jiang, and Gui Gui
|
Are Bus-Mounted Edge Servers Feasible?
| null | null | null | null |
cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Placement of edge servers is the prerequisite of provisioning edge computing services for Internet of Vehicles (IoV). Fixed-site edge servers at Road Side Units (RSUs) or base stations are able to offer basic service coverage for end users, i.e., vehicles on road. However, the server locations and capacity are fixed after deployment, rendering their inefficiency in handling spationtemporal user dynamics. Mobile servers such as buses, on the other hand, have the potential of adding computation elasticity to such system. To this end, this paper studies the feasibility of bus-mounted edge servers based on real traces. First, we investigate the coverage of the buses and base stations using the Shanghai bus/taxi/Telecom datasets, which shows a great potential of bus-based edge servers as they cover a great portion of geographic area and demand points. Next, we build a mathematical model and design a simple greedy heuristic algorithm to select a limited number of buses that maximizes the coverage of demand points, i.e., with a limited purchase budget. We perform trace-driven simulations to verify the performance of the proposed bus selection algorithm. The results show that our approach effectively handles the dynamic user demand under realistic constraints such as server capacity and purchase quantity. Thus, we claim: bus-mounted edge servers for vehicular networks in urban areas are feasible, beneficial, and valuable.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 08:56:15 GMT"
},
{
"version": "v2",
"created": "Mon, 08 Dec 2025 12:46:18 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 14:58:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Li",
"Xuezhi",
""
],
[
"He",
"Jiancong",
""
],
[
"Xie",
"Ming",
""
],
[
"Chen",
"Xuyang",
""
],
[
"Chang",
"Le",
""
],
[
"Jiang",
"Li",
""
],
[
"Gui",
"Gui",
""
]
] |
2512.05647
|
Konstantinos Skianis
|
Giorgos Antoniou, Giorgos Filandrianos, Aggelos Vlachos, Giorgos Stamou, Lampros Kollimenos, Konstantinos Skianis, Michalis Vazirgiannis
|
A Greek Government Decisions Dataset for Public-Sector Analysis and Insight
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce an open, machine-readable corpus of Greek government decisions sourced from the national transparency platform Diavgeia. The resource comprises 1 million decisions, featuring and high-quality raw text extracted from PDFs. It is released with raw extracted text in Markdown format, alongside a fully reproducible extraction pipeline. Beyond the core dataset, we conduct qualitative analyses to explore boilerplate patterns and design a retrieval-augmented generation (RAG) task by formulating a set of representative questions, creating high-quality answers, and evaluating a baseline RAG system on its ability to retrieve and reason over public decisions. This evaluation demonstrates the potential of large-scale public-sector corpora to support advanced information access and transparency through structured retrieval and reasoning over governmental documents, and highlights how such a RAG pipeline could simulate a chat-based assistant capable of interactively answering questions about public decisions. Due to its scale, quality, and domain coverage, the corpus can also serve as high-value pre-training or fine-tuning material for new Language Models (LMs) and Large Language Models (LLMs) respectively, including specialized models for legal and governmental domains, and as a foundation for novel approaches in domain adaptation, knowledge-grounded generation, and explainable AI. Finally, we discuss limitations, outline future directions, and make both the data and the code accessible.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 11:47:33 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 11:58:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Antoniou",
"Giorgos",
""
],
[
"Filandrianos",
"Giorgos",
""
],
[
"Vlachos",
"Aggelos",
""
],
[
"Stamou",
"Giorgos",
""
],
[
"Kollimenos",
"Lampros",
""
],
[
"Skianis",
"Konstantinos",
""
],
[
"Vazirgiannis",
"Michalis",
""
]
] |
2512.05832
|
Yifei Tong
|
Yifei Tong
|
Heard or Halted? Gender, Interruptions, and Emotional Tone in U.S. Supreme Court Oral Arguments
|
12 pages, 5 figures, 1 table. Includes appendix. Code available at: https://github.com/1TSHARUKA/Emotional_Interruption_Analysis
| null | null | null |
cs.CL cs.CY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This study examines how interruptions during U.S. Supreme Court oral arguments shape both the semantic content and emotional tone of advocates' speech, with a focus on gendered dynamics in judicial discourse. Using the ConvoKit Supreme Court Corpus (2010-2019), we analyze 12,663 speech chunks from advocate-justice interactions to assess whether interruptions alter the meaning of an advocate's argument and whether interruptions toward female advocates exhibit more negative emotional valence. Semantic shifts are quantified using GloVe-based sentence embeddings, while sentiment is measured through lexicon-based analysis. We find that semantic similarity between pre- and post-interruption speech remains consistently high, suggesting that interruptions do not substantially alter argumentative content. However, interruptions directed at female advocates contain significantly higher levels of negative sentiment. These results deepen empirical understanding of gendered communication in elite institutional settings and demonstrate the value of computational linguistic methods for studying power, discourse, and equity in judicial proceedings.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 15:56:17 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:30:02 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tong",
"Yifei",
""
]
] |
2512.05991
|
Xuanqi Zhou
|
Chang Liu, Tianjiao Jing, Chengcheng Ma, Xuanqi Zhou, Zhengxuan Lian, Qin Jin, Hongliang Yuan, Shi-Sheng Huang
|
EmoDiffTalk:Emotion-aware Diffusion for Editable 3D Gaussian Talking Head
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent photo-realistic 3D talking head via 3D Gaussian Splatting still has significant shortcoming in emotional expression manipulation, especially for fine-grained and expansive dynamics emotional editing using multi-modal control. This paper introduces a new editable 3D Gaussian talking head, i.e. EmoDiffTalk. Our key idea is a novel Emotion-aware Gaussian Diffusion, which includes an action unit (AU) prompt Gaussian diffusion process for fine-grained facial animator, and moreover an accurate text-to-AU emotion controller to provide accurate and expansive dynamic emotional editing using text input. Experiments on public EmoTalk3D and RenderMe-360 datasets demonstrate superior emotional subtlety, lip-sync fidelity, and controllability of our EmoDiffTalk over previous works, establishing a principled pathway toward high-quality, diffusion-driven, multimodal editable 3D talking-head synthesis. To our best knowledge, our EmoDiffTalk is one of the first few 3D Gaussian Splatting talking-head generation framework, especially supporting continuous, multimodal emotional editing within the AU-based expression space.
|
[
{
"version": "v1",
"created": "Sun, 30 Nov 2025 16:28:19 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 19:56:51 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liu",
"Chang",
""
],
[
"Jing",
"Tianjiao",
""
],
[
"Ma",
"Chengcheng",
""
],
[
"Zhou",
"Xuanqi",
""
],
[
"Lian",
"Zhengxuan",
""
],
[
"Jin",
"Qin",
""
],
[
"Yuan",
"Hongliang",
""
],
[
"Huang",
"Shi-Sheng",
""
]
] |
2512.06019
|
Aubrey Jaffer
|
Aubrey G. Jaffer and Martin S. Jaffer
|
Natural Convection Heat Transfer from an Inclined Cylinder
|
15 pages; 8 figures; 7 tables; 17 references
| null | null | null |
physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This investigation derives a novel formula predicting the natural convective heat transfer from an inclined cylinder given its length, diameter, inclination angle, Rayleigh number, and the fluid's Prandtl number and thermal conductivity.
The present formula was tested with 93 inclined cylinder measurements having length-to-diameter ratios between 1.48 and 104 in nine data-sets from three peer-reviewed studies, yielding (data-set) root-mean-squared relative error values between 1.6% and 4.7%.
|
[
{
"version": "v1",
"created": "Thu, 04 Dec 2025 02:41:37 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 22:24:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jaffer",
"Aubrey G.",
""
],
[
"Jaffer",
"Martin S.",
""
]
] |
2512.06112
|
Yifang Xu
|
Yifang Xu, Jiahao Cui, Feipeng Cai, Zhihao Zhu, Hanlin Shang, Shan Luan, Mingwang Xu, Neng Zhang, Yaoyi Li, Jia Cai, Siyu Zhu
|
WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving
|
18 pages, 11 figures. Code & Model: https://github.com/fudan-generative-vision/WAM-Flow
| null | null | null |
cs.RO cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 89.1 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving. The code will be publicly available soon.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 19:36:46 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:06:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Yifang",
""
],
[
"Cui",
"Jiahao",
""
],
[
"Cai",
"Feipeng",
""
],
[
"Zhu",
"Zhihao",
""
],
[
"Shang",
"Hanlin",
""
],
[
"Luan",
"Shan",
""
],
[
"Xu",
"Mingwang",
""
],
[
"Zhang",
"Neng",
""
],
[
"Li",
"Yaoyi",
""
],
[
"Cai",
"Jia",
""
],
[
"Zhu",
"Siyu",
""
]
] |
2512.06181
|
Yanuo Zhou
|
Yanuo Zhou
|
Beyond Lux thresholds: a systematic pipeline for classifying biologically relevant light contexts from wearable data
|
Withdrawn at the request of affiliated institution
| null | null | null |
q-bio.QM cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Background: Wearable spectrometers enable field quantification of biologically relevant light, yet reproducible pipelines for contextual classification remain under-specified.
Objective: To establish and validate a subject-wise evaluated, reproducible pipeline and actionable design rules for classifying natural vs. artificial light from wearable spectral data.
Methods: We analysed ActLumus recordings from 26 participants, each monitored for at least 7 days at 10-second sampling, paired with daily exposure diaries. The pipeline fixes the sequence: domain selection, log-base-10 transform, L2 normalisation excluding total intensity (to avoid brightness shortcuts), hour-level medoid aggregation, sine/cosine hour encoding, and MLP classifier, evaluated under participant-wise cross-validation.
Results: The proposed sequence consistently achieved high performance on the primary task, with representative configurations reaching AUC = 0.938 (accuracy 88%) for natural vs. artificial classification on the held-out subject split. In contrast, indoor vs. outdoor classification remained at feasibility level due to spectral overlap and class imbalance (best AUC approximately 0.75; majority-class collapse without contextual sensors). Threshold baselines were insufficient on our data, supporting the need for spectral-temporal modelling beyond illuminance cut-offs.
Conclusions: We provide a reproducible, auditable baseline pipeline and design rules for contextual light classification under subject-wise generalisation. All code, configuration files, and derived artefacts will be openly archived (GitHub + Zenodo DOI) to support reuse and benchmarking.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 22:02:02 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 07:50:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhou",
"Yanuo",
""
]
] |
2512.06356
|
Yifan Song
|
Yifan Song, Fenglin Yu, Yihong Luo, Xingjian Tao, Siya Qiu, Kai Han, Jing Tang
|
DDFI: Diverse and Distribution-aware Missing Feature Imputation via Two-step Reconstruction
| null | null | null | null |
cs.LG cs.SI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Incomplete node features are ubiquitous in real-world scenarios, e.g., the attributes of web users may be partly private, which causes the performance of Graph Neural Networks (GNNs) to decline significantly. Feature propagation (FP) is a well-known method that performs well for imputation of missing node features on graphs, but it still has the following three issues: 1) it struggles with graphs that are not fully connected, 2) imputed features face the over-smoothing problem, and 3) FP is tailored for transductive tasks, overlooking the feature distribution shift in inductive tasks. To address these challenges, we introduce DDFI, a Diverse and Distribution-aware Missing Feature Imputation method that combines feature propagation with a graph-based Masked AutoEncoder (MAE) in a nontrivial manner. It first designs a simple yet effective algorithm, namely Co-Label Linking (CLL), that randomly connects nodes in the training set with the same label to enhance the performance on graphs with numerous connected components. Then we develop a novel two-step representation generation process at the inference stage. Specifically, instead of directly using FP-imputed features as input during inference, DDFI further reconstructs the features through the whole MAE to reduce feature distribution shift in the inductive tasks and enhance the diversity of node features. Meanwhile, since existing feature imputation methods for graphs only evaluate by simulating the missing scenes with manually masking the features, we collect a new dataset called Sailing from the records of voyages that contains naturally missing features to help better evaluate the effectiveness. Extensive experiments conducted on six public datasets and Sailing show that DDFI outperforms the state-of-the-art methods under both transductive and inductive settings.
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 09:06:08 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:53:17 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Song",
"Yifan",
""
],
[
"Yu",
"Fenglin",
""
],
[
"Luo",
"Yihong",
""
],
[
"Tao",
"Xingjian",
""
],
[
"Qiu",
"Siya",
""
],
[
"Han",
"Kai",
""
],
[
"Tang",
"Jing",
""
]
] |
2512.06364
|
Shyama Sastha Krishnamoorthy Srinivasan
|
Shyama Sastha Krishnamoorthy Srinivasan, Harsh Pala, Mohan Kumar, Pushpendra Singh
|
JEEVHITAA -- An End-to-End HCAI System to Support Collective Care
|
14 pages, 4 figures, 6 tables, 2 algorithms, Preprint of ongoing work. Longer Pilot work underway
| null | null | null |
cs.CR cs.HC
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Current mobile health platforms are predominantly individual-centric and lack the necessary primitives for coordinated, auditable, multi-actor workflows. However, in many settings worldwide, health decisions are enacted by multi-actor care networks rather than single users. We present JEEVHITAA, an Android/Flutter system that provides context-sensitive, role-aware sharing and verifiable information flows for care circles. JEEVHITAA ingests platform and device data (via Google Health Connect and BLE connectors), constructs multi-layer user profiles from sensor streams and tiered onboarding, and enforces fine-grained, time-bounded access control across permissioned care graphs. Data are end-to-end encrypted in local stores and during peer sync (Firebase), and provisions are made for document capture by camera or upload as PDF. An integrated retrieval-augmented LLM pipeline (i) produces structured, role-targeted summaries and action plans, (ii) enables users to gather advanced insights on health reports, and (iii) performs evidence-grounded user-relevant verification of arbitrary health content, returning provenance, confidence scores, and source citations. We describe the system architecture, connector abstractions, and security primitives, and evaluate robustness and compatibility using synthetic, ontology-driven simulations and vendor compatibility tests. Finally, we outline plans for longitudinal in-the-wild deployments to measure system performance, the correctness of access control, and the real-world effectiveness of relationship-aware credibility support.
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 09:36:27 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 17:26:03 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Srinivasan",
"Shyama Sastha Krishnamoorthy",
""
],
[
"Pala",
"Harsh",
""
],
[
"Kumar",
"Mohan",
""
],
[
"Singh",
"Pushpendra",
""
]
] |
2512.06392
|
Lefan Zhang
|
Runlong Zhou, Lefan Zhang, Shang-Chen Wu, Kelvin Zou, Hanzhi Zhou, Ke Ye, Yihao Feng, Dong Yin, Alex Guillen Garcia, Dmytro Babych, Rohit Chatterjee, Matthew Hopkins, Xiang Kong, Chang Lan, Lezhi Li, Yiping Ma, Daniele Molinari, Senyu Tong, Yanchao Sun, Thomas Voice, Jianyu Wang, Chong Wang, Simon Wang, Floris Weers, Yechen Xu, Guolin Yin, Muyang Yu, Yi Zhang, Zheng Zhou, Danyang Zhuo, Ruoming Pang, Cheng Leong
|
RLAX: Large-Scale, Distributed Reinforcement Learning for Large Language Models on TPUs
|
The submission is being withdrawn because internal stakeholders determined that it is not appropriate to publish work on this topic at this time
| null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reinforcement learning (RL) has emerged as the de-facto paradigm for improving the reasoning capabilities of large language models (LLMs). We have developed RLAX, a scalable RL framework on TPUs. RLAX employs a parameter-server architecture. A master trainer periodically pushes updated model weights to the parameter server while a fleet of inference workers pull the latest weights and generates new rollouts. We introduce a suite of system techniques to enable scalable and preemptible RL for a diverse set of state-of-art RL algorithms. To accelerate convergence and improve model quality, we have devised new dataset curation and alignment techniques. Large-scale evaluations show that RLAX improves QwQ-32B's pass@8 accuracy by 12.8% in just 12 hours 48 minutes on 1024 v5p TPUs, while remaining robust to preemptions during training.
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 10:48:51 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:38:26 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhou",
"Runlong",
""
],
[
"Zhang",
"Lefan",
""
],
[
"Wu",
"Shang-Chen",
""
],
[
"Zou",
"Kelvin",
""
],
[
"Zhou",
"Hanzhi",
""
],
[
"Ye",
"Ke",
""
],
[
"Feng",
"Yihao",
""
],
[
"Yin",
"Dong",
""
],
[
"Garcia",
"Alex Guillen",
""
],
[
"Babych",
"Dmytro",
""
],
[
"Chatterjee",
"Rohit",
""
],
[
"Hopkins",
"Matthew",
""
],
[
"Kong",
"Xiang",
""
],
[
"Lan",
"Chang",
""
],
[
"Li",
"Lezhi",
""
],
[
"Ma",
"Yiping",
""
],
[
"Molinari",
"Daniele",
""
],
[
"Tong",
"Senyu",
""
],
[
"Sun",
"Yanchao",
""
],
[
"Voice",
"Thomas",
""
],
[
"Wang",
"Jianyu",
""
],
[
"Wang",
"Chong",
""
],
[
"Wang",
"Simon",
""
],
[
"Weers",
"Floris",
""
],
[
"Xu",
"Yechen",
""
],
[
"Yin",
"Guolin",
""
],
[
"Yu",
"Muyang",
""
],
[
"Zhang",
"Yi",
""
],
[
"Zhou",
"Zheng",
""
],
[
"Zhuo",
"Danyang",
""
],
[
"Pang",
"Ruoming",
""
],
[
"Leong",
"Cheng",
""
]
] |
2512.06438
|
Ramazan Fazylov
|
Ramazan Fazylov, Sergey Zagoruyko, Aleksandr Parkin, Stamatis Lefkimmiatis, Ivan Laptev
|
AGORA: Adversarial Generation Of Real-time Animatable 3D Gaussian Head Avatars
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
The generation of high-fidelity, animatable 3D human avatars remains a core challenge in computer graphics and vision, with applications in VR, telepresence, and entertainment. Existing approaches based on implicit representations like NeRFs suffer from slow rendering and dynamic inconsistencies, while 3D Gaussian Splatting (3DGS) methods are typically limited to static head generation, lacking dynamic control. We bridge this gap by introducing AGORA, a novel framework that extends 3DGS within a generative adversarial network to produce animatable avatars. Our key contribution is a lightweight, FLAME-conditioned deformation branch that predicts per-Gaussian residuals, enabling identity-preserving, fine-grained expression control while allowing real-time inference. Expression fidelity is enforced via a dual-discriminator training scheme leveraging synthetic renderings of the parametric mesh. AGORA generates avatars that are not only visually realistic but also precisely controllable. Quantitatively, we outperform state-of-the-art NeRF-based methods on expression accuracy while rendering at 250+ FPS on a single GPU, and, notably, at $\sim$9 FPS under CPU-only inference - representing, to our knowledge, the first demonstration of practical CPU-only animatable 3DGS avatar synthesis. This work represents a significant step toward practical, high-performance digital humans. Project website: https://ramazan793.github.io/AGORA/
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 14:05:20 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 19:38:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fazylov",
"Ramazan",
""
],
[
"Zagoruyko",
"Sergey",
""
],
[
"Parkin",
"Aleksandr",
""
],
[
"Lefkimmiatis",
"Stamatis",
""
],
[
"Laptev",
"Ivan",
""
]
] |
2512.06548
|
Hyeonwoo Nam
|
Hyeon Woo Nam, Tae Woong Jeong, and Sung Min Jo
|
An Euler-Lagrangian Multiphysics Coupling Framework for Particle-Laden High-Speed Flows
|
24 pages, 11 figures
| null | null | null |
physics.flu-dyn physics.app-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Particle-laden effects in high-speed flows require a coupled Euler and Lagrangian prediction technique with varying fidelity of thermochemical models, depending on the simulation conditions of interest. This requirement makes the development of a conventional monolithic solver challenging to manage the different fidelity of the thermochemical models within a single computational framework. To address this, the present study proposes a multi-solver framework for the coupled Euler-Lagrangian predictions applicable to various particle-laden high-speed flow conditions. Volumetric and surface couplings are established between a particle solver ORACLE (OpenFOAM-based lagRAngian CoupLEr) and a thermochemical nonequilibrium flow solver based on an adaptable data exchange algorithm. The developed framework is then validated by predicting particle-laden supersonic nozzle flows and aerothermal heating around a hypersonic Martian atmospheric entry capsule. Finally, a quasi-1D approximation is proposed in conjunction with a surrogate method to efficiently and accurately predict particle-laden surface erosion, with quantified parametric uncertainty, for hypersonic aerothermal characterization.
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 19:41:04 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 00:52:17 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 08:17:37 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nam",
"Hyeon Woo",
""
],
[
"Jeong",
"Tae Woong",
""
],
[
"Jo",
"Sung Min",
""
]
] |
2512.06571
|
Zifan Xu
|
Zifan Xu, Myoungkyu Seo, Dongmyeong Lee, Hao Fu, Jiaheng Hu, Jiaxun Cui, Yuqian Jiang, Zhihan Wang, Anastasiia Brund, Joydeep Biswas, Peter Stone
|
Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input
| null | null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Learning fast and robust ball-kicking skills is a critical capability for humanoid soccer robots, yet it remains a challenging problem due to the need for rapid leg swings, postural stability on a single support foot, and robustness under noisy sensory input and external perturbations (e.g., opponents). This paper presents a reinforcement learning (RL)-based system that enables humanoid robots to execute robust continual ball-kicking with adaptability to different ball-goal configurations. The system extends a typical teacher-student training framework -- in which a "teacher" policy is trained with ground truth state information and the "student" learns to mimic it with noisy, imperfect sensing -- by including four training stages: (1) long-distance ball chasing (teacher); (2) directional kicking (teacher); (3) teacher policy distillation (student); and (4) student adaptation and refinement (student). Key design elements -- including tailored reward functions, realistic noise modeling, and online constrained RL for adaptation and refinement -- are critical for closing the sim-to-real gap and sustaining performance under perceptual uncertainty. Extensive evaluations in both simulation and on a real robot demonstrate strong kicking accuracy and goal-scoring success across diverse ball-goal configurations. Ablation studies further highlight the necessity of the constrained RL, noise modeling, and the adaptation stage. This work presents a system for learning robust continual humanoid ball-kicking under imperfect perception, establishing a benchmark task for visuomotor skill learning in humanoid whole-body control.
|
[
{
"version": "v1",
"created": "Sat, 06 Dec 2025 21:27:50 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:53:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Zifan",
""
],
[
"Seo",
"Myoungkyu",
""
],
[
"Lee",
"Dongmyeong",
""
],
[
"Fu",
"Hao",
""
],
[
"Hu",
"Jiaheng",
""
],
[
"Cui",
"Jiaxun",
""
],
[
"Jiang",
"Yuqian",
""
],
[
"Wang",
"Zhihan",
""
],
[
"Brund",
"Anastasiia",
""
],
[
"Biswas",
"Joydeep",
""
],
[
"Stone",
"Peter",
""
]
] |
2512.06613
|
Yueying Ke
|
Yueying Ke
|
Hierarchical Deep Learning for Diatom Image Classification: A Multi-Level Taxonomic Approach
|
Version 2: Corrected reference details, improved architectural diagram, and enhanced writing for clarity and precision. Added a table illustrating the masking mechanism. No changes to experimental results or conclusions. 11 pages, 6 figures, 3 tables
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Accurate taxonomic identification of diatoms is essential for aquatic ecosystem monitoring, yet conventional methods depend heavily on expert taxonomists. Recent deep learning approaches improve automation, but most treat diatom recognition as flat classification, predicting only one taxonomic rank. We investigate whether embedding taxonomic hierarchy into neural network architectures can improve both accuracy and error locality.
We introduce DiatomCascadeNet (H-COFGS), a hierarchical convolutional network with five cascaded heads that jointly predict class, order, family, genus, and species. Each head receives shared backbone features and probability distributions from higher levels, with binary masks restricting predictions to valid descendants during training and inference. Using a filtered dataset of 1,456 diatom images covering 82 species, we compare hierarchical and flat models under identical settings.
H-COFGS matches flat baselines at the species level (69.4% accuracy) while outperforming at all upper taxonomic levels. When species predictions fail, errors remain taxonomically local: 92.5% of misclassified species are correctly predicted at the genus level, versus 67.2% for flat baselines. H-COFGS reduces mean taxonomic distance by 38.2% (1.209 vs. 1.955).
Progressive training reveals bidirectional mechanisms: hierarchical constraint masks operate top-down to constrain prediction space, while gradients from fine-grained levels propagate bottom-up through the shared backbone, refining features. This improves class accuracy from 96.2% to 99.5% and yields 6-8% gains at upper levels, producing more robust, interpretable, and biologically aligned predictions for multi-level taxonomic classification.
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 01:06:13 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:51:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ke",
"Yueying",
""
]
] |
2512.06891
|
Huaijin Ma
|
Huaijin Ma, Te Liu, Jiachen Sheng, Kaiyan Cao, Jinpeng Yang, Jian Wang
|
Observation of Stable Bimeron Transport Driven by Spoof Surface Acoustic Waves on Chiral Metastructures
|
17 pages, 7 figures
| null | null | null |
cond-mat.mtrl-sci physics.app-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Topological quasiparticles, such as merons and bimerons, are characterized by non-trivial textures that exhibit remarkably robust transport against deformation, offering significant potential for information processing. While these phenomena have been explored in various systems, acoustic realizations remain challenging. Here, we report that acoustic meron topological textures were successfully realized using designed Archimedeanlike square spiral metastructures via the excitation of spoof surface acoustic waves (SSAWs). By applying mirror-symmetric combinatorial operations to the unit structures, we further construct composite chiral metastructures that enable both one-dimensional and two-dimensional stable transport of acoustic bimerons. It is further revealed that bimeron transport originates from the locked opposite phase differences of SSAWs, induced by the handedness of the cavity resonant modes. The intrinsic robustness of the meron textures against structural defects is confirmed through the calculation of their topological charge. Our findings establish stable acoustic bimeron transport as a topologically resilient foundation for future acoustic information processing and storage technologies.
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 15:34:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ma",
"Huaijin",
""
],
[
"Liu",
"Te",
""
],
[
"Sheng",
"Jiachen",
""
],
[
"Cao",
"Kaiyan",
""
],
[
"Yang",
"Jinpeng",
""
],
[
"Wang",
"Jian",
""
]
] |
2512.06918
|
Da Xu
|
Da Xu
|
Angular Momentum Penrose Inequality
| null | null | null | null |
gr-qc math.FA
|
http://creativecommons.org/licenses/by/4.0/
|
We prove the Angular Momentum Penrose Inequality, a fundamental result connecting the total mass of an isolated gravitational system to the size and spin of any black holes it contains. The inequality states that the mass of a spacetime must be at least as large as a specific combination of the black hole's horizon area and angular momentum, with the bound being tight precisely for the Kerr family of rotating black holes.
The proof combines four techniques: solving a geometric equation that straightens out the initial data, applying a conformal transformation that encodes angular momentum, tracking how angular momentum is preserved through the construction, and invoking known bounds that prevent black holes from spinning too fast. A central innovation is a new notion of mass that incorporates both the standard Hawking mass and angular momentum, and which increases monotonically along a natural geometric flow from the black hole horizon out to infinity.
As an application, we also prove the Charged Penrose Inequality for non-rotating charged black holes, showing that electric charge contributes to the mass bound in a manner analogous to angular momentum.
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 16:53:02 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:29:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Da",
""
]
] |
2512.06982
|
Qian Xie
|
Yu Yu, Qian Xie, Nairen Cao, Li Jin
|
LLM-Driven Composite Neural Architecture Search for Multi-Source RL State Encoding
|
NeurIPS 2025 Workshop on Bridging Language, Agent, and World Models for Reasoning and Planning
| null | null | null |
cs.LG cs.SY eess.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Designing state encoders for reinforcement learning (RL) with multiple information sources -- such as sensor measurements, time-series signals, image observations, and textual instructions -- remains underexplored and often requires manual design. We formalize this challenge as a problem of composite neural architecture search (NAS), where multiple source-specific modules and a fusion module are jointly optimized. Existing NAS methods overlook useful side information from the intermediate outputs of these modules -- such as their representation quality -- limiting sample efficiency in multi-source RL settings. To address this, we propose an LLM-driven NAS pipeline in which the LLM serves as a neural architecture design agent, leveraging language-model priors and intermediate-output signals to guide sample-efficient search for high-performing composite state encoders. On a mixed-autonomy traffic control task, our approach discovers higher-performing architectures with fewer candidate evaluations than traditional NAS baselines and the LLM-based GENIUS framework.
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 20:25:07 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:52:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yu",
"Yu",
""
],
[
"Xie",
"Qian",
""
],
[
"Cao",
"Nairen",
""
],
[
"Jin",
"Li",
""
]
] |
2512.07019
|
Yuqi Gu
|
Zhiyu Xu, Jia Liu, Yixin Wang, Yuqi Gu
|
Latency-Response Theory Model: Evaluating Large Language Models via Response Accuracy and Chain-of-Thought Length
| null | null | null | null |
stat.ME cs.AI stat.AP stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The proliferation of Large Language Models (LLMs) necessitates valid evaluation methods to guide downstream applications and actionable future improvements. The Item Response Theory (IRT) has recently emerged as a promising framework for evaluating LLMs via their response accuracy. Beyond simple response accuracy, LLMs' chain of thought (CoT) lengths serve as a vital indicator of their reasoning ability. To leverage the CoT length information to assist the evaluation of LLMs, we propose Latency-Response Theory (LaRT) to jointly model the response accuracy and CoT length by introducing the latent ability, latent speed, and a key correlation parameter between them. We derive an efficient estimation algorithm and establish rigorous identifiability results for the population parameters to ensure the statistical validity of estimation. Theoretical asymptotic analyses and simulation studies demonstrate LaRT's advantages over IRT in terms of higher estimation accuracy and shorter confidence intervals for latent traits. A key finding is that the asymptotic estimation precision of the latent ability under LaRT exceeds that of IRT whenever the latent ability and latent speed are correlated. We collect real responses from diverse LLMs on popular benchmark datasets. The application of LaRT reveals a strong negative correlation between the latent ability and latent speed in all benchmarks, with stronger correlation for more difficult benchmarks. This finding supports the intuition that higher reasoning ability correlates with slower speed and longer response latency. LaRT yields different LLM rankings than IRT and outperforms IRT across multiple key evaluation metrics including predictive power, item efficiency, ranking validity, and LLM evaluation efficiency. Code and data are available at https://github.com/Toby-X/Latency-Response-Theory-Model.
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 22:06:51 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:45:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Zhiyu",
""
],
[
"Liu",
"Jia",
""
],
[
"Wang",
"Yixin",
""
],
[
"Gu",
"Yuqi",
""
]
] |
2512.07031
|
Anton Shafarevich
|
Anton Shafarevich
|
Flexibility of affine spherical varieties
| null | null | null | null |
math.AG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove that the automorphism group $\mathrm{Aut}(X)$ of an affine spherical variety $X$ acts transitively on the set of smooth points $X^{reg}.$ If every invertible regular function on $X$ is constant, we prove that $X$ is flexible, i.e., the subgroup of $\mathrm{Aut}(X)$ generated by all $\mathbb{G}_a$-subgroups acts transitively on $X^{reg}.$
|
[
{
"version": "v1",
"created": "Sun, 07 Dec 2025 22:48:08 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:43:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shafarevich",
"Anton",
""
]
] |
2512.07062
|
Changliang Xia
|
Changliang Xia, Chengyou Jia, Minnan Luo, Zhuohang Dang, Xin Shen, Bowen Ping
|
$\mathrm{D}^\mathrm{3}$-Predictor: Noise-Free Deterministic Diffusion for Dense Prediction
| null | null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Although diffusion models with strong visual priors have emerged as powerful dense prediction backboens, they overlook a core limitation: the stochastic noise at the core of diffusion sampling is inherently misaligned with dense prediction that requires a deterministic mapping from image to geometry. In this paper, we show that this stochastic noise corrupts fine-grained spatial cues and pushes the model toward timestep-specific noise objectives, consequently destroying meaningful geometric structure mappings. To address this, we introduce $\mathrm{D}^\mathrm{3}$-Predictor, a noise-free deterministic framework built by reformulating a pretrained diffusion model without stochasticity noise. Instead of relying on noisy inputs to leverage diffusion priors, $\mathrm{D}^\mathrm{3}$-Predictor views the pretrained diffusion network as an ensemble of timestep-dependent visual experts and self-supervisedly aggregates their heterogeneous priors into a single, clean, and complete geometric prior. Meanwhile, we utilize task-specific supervision to seamlessly adapt this noise-free prior to dense prediction tasks. Extensive experiments on various dense prediction tasks demonstrate that $\mathrm{D}^\mathrm{3}$-Predictor achieves competitive or state-of-the-art performance in diverse scenarios. In addition, it requires less than half the training data previously used and efficiently performs inference in a single step. Our code, data, and checkpoints are publicly available at https://x-gengroup.github.io/HomePage_D3-Predictor/.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 00:39:32 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:07:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xia",
"Changliang",
""
],
[
"Jia",
"Chengyou",
""
],
[
"Luo",
"Minnan",
""
],
[
"Dang",
"Zhuohang",
""
],
[
"Shen",
"Xin",
""
],
[
"Ping",
"Bowen",
""
]
] |
2512.07098
|
Samuel Goodman
|
Samuel Goodman
|
Regular Functions on Formal-Analytic Arithmetic Surfaces
|
11 pages
| null | null | null |
math.CV math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we show that for a broad class of pseudoconvex formal-analytic arithmetic surfaces over $\text{Spec}(\mathbb{Z})$, those which admit a nonconstant monic such regular function, that a conjecture of Bost-Charles that the ring of regular functions has continuum cardinality is implied by a purely complex-analytic conjecture. Under the conjecture, a Fekete-Szego-type approximation argument produces a polynomial "large" relative to the regular function, which in turn yields continuum many distinct regular functions. We also introduce a formula for the pushforward by a holomorphic function of the equilibrium Green's functions for our bordered Riemann surface with boundary, a formula which has constant term related to Arakelov degree.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 02:27:57 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:08:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Goodman",
"Samuel",
""
]
] |
2512.07121
|
Megan Brown
|
Megan A. Brown, Tiago Ventura, Joshua A. Tucker, Jonathan Nagler
|
The relationship between offline partisan geographical segregation and online partisan segregation
| null | null | null | null |
cs.SI cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
Social media is often blamed for the creation of echo chambers. However, these claims fail to consider the prevalence of offline echo chambers resulting from high levels of partisan segregation in the United States. Our article empirically assesses these online versus offline dynamics by linking a novel dataset of voters' offline partisan segregation extracted from publicly available voter files for 180 million US voters with their online network segregation on Twitter. We investigate offline and online partisan segregation using measures of geographical and network isolation of every matched voter-twitter user to their co-partisans online and offline. Our results show that while social media users tend to form politically homogeneous online networks, these levels of partisan sorting are significantly lower than those found in offline settings. Notably, Democrats are more isolated than Republicans in both settings, and only older Republicans exhibit higher online than offline segregation. Our results contribute to the emerging literature on political communication and the homophily of online networks, providing novel evidence on partisan sorting both online and offline.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 03:04:11 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:41:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Brown",
"Megan A.",
""
],
[
"Ventura",
"Tiago",
""
],
[
"Tucker",
"Joshua A.",
""
],
[
"Nagler",
"Jonathan",
""
]
] |
2512.07321
|
Marlo Kunzner
|
Marlo Kunzner, Luis Henriques, Fahad Puthalath, Leonardo Facchini, Mohammadhossein Shahsavari, L\'ea Gommeringer, Martin Angelmahr, Peidong Yu, Matthias Sperl, Till B\"ohmer, and Jan Philipp Gabriel
|
Towards probing velocity distributions in dense granular matter: Utilizing Fiber Bragg Gratings
| null | null | null | null |
cond-mat.dis-nn cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Granular gases are commonly characterized through their velocity distribution, which provides access to the granular temperature. In experiments, velocity distributions are typically obtained by particle tracking, which however becomes limited at moderate and high particle densities. As a way forward, we propose a new technique for measuring particle velocities in situ by using a Fiber Bragg Grating (FBG) sensor, which remains applicable at significantly higher particle densities.The FBG sensor detects strain pulses induced by particle-fiber collisions, from which the velocity of the impacting particle can be derived. Applying this method to an ensemble of granular particles allows to extract its velocity distributions as we present for a granular system excited by a vibrational shaker. We validate the extracted velocity distribution against conventional particle-tracking measurements, confirming the reliability of the FBG-based technique.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 09:04:18 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:31:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kunzner",
"Marlo",
""
],
[
"Henriques",
"Luis",
""
],
[
"Puthalath",
"Fahad",
""
],
[
"Facchini",
"Leonardo",
""
],
[
"Shahsavari",
"Mohammadhossein",
""
],
[
"Gommeringer",
"Léa",
""
],
[
"Angelmahr",
"Martin",
""
],
[
"Yu",
"Peidong",
""
],
[
"Sperl",
"Matthias",
""
],
[
"Böhmer",
"Till",
""
],
[
"Gabriel",
"Jan Philipp",
""
]
] |
2512.07364
|
Handhika Ramadhan
|
S. M. Holme, H. S. Ramadhan, I. Nurul Huda, and Leonardus B. Putra
|
Orbital dynamics and spin-precession around a circular chiral vorton
|
36 pages, 22 figures
| null | null | null |
gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
Vortons are of interest in high-energy physics as possible dark matter candidates and as probes of Grand Unified Theories. Using the recently derived weak-field metric for a chiral vorton, we study the dynamics of test particles by analyzing both timelike and null geodesics. We identify several classes of trajectories, including bound precessing orbits, circular orbits, toroidal, and crown-type oscillations, as well as unbound scattering paths. Poincare surfaces of section reveal transitions between regular and chaotic motions that depend sensitively on the vorton tension $G\mu$ and initial conditions. We further compute the Lense-Thirring and general spin-precession frequencies for gyroscopes along Killing trajectories. The resulting precession profiles exhibit several distinct features not present in Kerr black holes but reminiscent of Kerr naked singularities, such as: divergences near the ring core, and multi-minima structures. These dynamical and precessional signatures may offer potential observational pathways for detecting vortons.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 10:03:05 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:51:41 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Holme",
"S. M.",
""
],
[
"Ramadhan",
"H. S.",
""
],
[
"Huda",
"I. Nurul",
""
],
[
"Putra",
"Leonardus B.",
""
]
] |
2512.07421
|
Wei-Hong Liang
|
Pei-Sen Su, Wen-Tao Lyu, Wei-Hong Liang, Eulogio Oset
|
The $D_{s0}^*(2317)^+$ decay to $D_s^+\pi^0$ and $D_s^{*+}\gamma$
|
12 pages, 7 figures, 3 tables; V2: new references added
| null | null | null |
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We study the strong decay of $D_{s0}^*(2317)^+$ to $D_s^+ \pi^0$ considering the coupled channels of $D^0 K^+, D^+ K^0, D_s^+ \eta$ and $D_s^+ \pi^0$ within a picture for the interaction based on the local hidden gauge approach. We also address the problem of the radiative decay to $D_s^{*+} \gamma$, using the same information obtained from the molecular picture. We obtain a strong width of the $D_{s0}^*(2317)^+$ of about $71 \,\rm keV$ and a radiative decay of $1.3 \,\rm keV$. The anomalous terms for the radiative decay are considered for the first time, but they are found negligible. We make a thorough discussion of this and other results to the light of the recent measurement of Belle for the ratio of these two decay modes, and make a call for the precise measurement of the two decay widths independently to clarify the present situation concerning the nature of of the $D_{s0}^*(2317)^+$ state.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 10:53:07 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:35:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Su",
"Pei-Sen",
""
],
[
"Lyu",
"Wen-Tao",
""
],
[
"Liang",
"Wei-Hong",
""
],
[
"Oset",
"Eulogio",
""
]
] |
2512.07524
|
Yunhao Qiu
|
Yunhao Qiu and Qinghai Zhang
|
A linear MARS method for three-dimensional interface tracking
| null | null | null | null |
math.NA cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For explicit interface tracking in three dimensions, we propose a linear MARS method that (a) represents the interface by a partially ordered set of glued surfaces and approximates each glued surface with a triangular mesh, (b) maintains an $(r,h,\theta)$-regularity on each triangular mesh so that the distance between any pair of adjacent markers is within the range $[rh,h]$ and no angle in any triangle is less than $\theta$, (c) applies to three-dimensional continua with arbitrarily complex topology and geometry, (d) preserves topological structures and geometric features of moving phases under diffeomorphic and isometric flow maps, and (e) achieves second-order and third-order accuracy in terms of the Lagrangian and Eulerian length scales, respectively. Results of classic benchmark tests verify the effectiveness of the novel mesh adjustment algorithms in enforcing the $(r,h,\theta)$-regularity and demonstrate the high accuracy and efficiency of the proposed linear MARS method.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 12:59:40 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:28:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Qiu",
"Yunhao",
""
],
[
"Zhang",
"Qinghai",
""
]
] |
2512.07565
|
Eugene Vasiliev O
|
E.O. Vasiliev, S.A. Drozdov, P.V. Baklanov, O.P. Vorobyov, S.Yu. Dedikov, M.S. Kirsanova, T.I. Larchenkova, N.N. Shakhvorostova
|
Interstellar Medium in Extremely High Star-Formation Regions: A Prospect of Observations on the Millimetron Space Observatory
|
22 pages, 7 figures, published in Astronomy Reports
|
Astronomy Reports, v. 69, pp. 913-929 (2025)
|
10.1134/S1063772925702154
| null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
High star-formation rate and active galactic nucleus' emission can significantly transform the interstellar medium. In ultra-luminous infrared galaxies, in which the star-formation rate reaches thousands of solar masses per year, the gas and dust are considerably affected by the ionizing radiation, cosmic rays and shock waves, that can be about a factor of 100--1000 larger than typical values in quiet star-forming galaxies. In these conditions, the emissivity of the gas and dust changes: in dense gas, high ionic and molecular transitions become excited, while dust grains are heated to high temperatures. In this paper, we analyze the possibilities for studying the interstellar medium in extreme conditions of ultra-luminous infrared galaxies at redshifts of $\sim 0-3$, utilizing the atomic and molecular lines, and dust continuum in far infrared range of $100-500\mu$m. We discuss the prospect of observations using the instruments of the Millimetron Space Observatory.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 13:58:55 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 19:18:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Vasiliev",
"E. O.",
""
],
[
"Drozdov",
"S. A.",
""
],
[
"Baklanov",
"P. V.",
""
],
[
"Vorobyov",
"O. P.",
""
],
[
"Dedikov",
"S. Yu.",
""
],
[
"Kirsanova",
"M. S.",
""
],
[
"Larchenkova",
"T. I.",
""
],
[
"Shakhvorostova",
"N. N.",
""
]
] |
2512.07602
|
Pengfei Sun
|
Pengfei Sun and Zhe Su and Jascha Achterberg and Giacomo Indiveri and Dan F.M. Goodman and Danyal Akarca
|
Algorithm-hardware co-design of neuromorphic networks with dual memory pathways
| null | null | null | null |
cs.NE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Spiking neural networks excel at event-driven sensing. Yet, maintaining task-relevant context over long timescales both algorithmically and in hardware, while respecting both tight energy and memory budgets, remains a core challenge in the field. We address this challenge through novel algorithm-hardware co-design effort. At the algorithm level, inspired by the cortical fast-slow organization in the brain, we introduce a neural network with an explicit slow memory pathway that, combined with fast spiking activity, enables a dual memory pathway (DMP) architecture in which each layer maintains a compact low-dimensional state that summarizes recent activity and modulates spiking dynamics. This explicit memory stabilizes learning while preserving event-driven sparsity, achieving competitive accuracy on long-sequence benchmarks with 40-60% fewer parameters than equivalent state-of-the-art spiking neural networks. At the hardware level, we introduce a near-memory-compute architecture that fully leverages the advantages of the DMP architecture by retaining its compact shared state while optimizing dataflow, across heterogeneous sparse-spike and dense-memory pathways. We show experimental results that demonstrate more than a 4x increase in throughput and over a 5x improvement in energy efficiency compared with state-of-the-art implementations. Together, these contributions demonstrate that biological principles can guide functional abstractions that are both algorithmically effective and hardware-efficient, establishing a scalable co-design paradigm for real-time neuromorphic computation and learning.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 14:50:26 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:28:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Sun",
"Pengfei",
""
],
[
"Su",
"Zhe",
""
],
[
"Achterberg",
"Jascha",
""
],
[
"Indiveri",
"Giacomo",
""
],
[
"Goodman",
"Dan F. M.",
""
],
[
"Akarca",
"Danyal",
""
]
] |
2512.07654
|
Boaz Moerman
|
Boaz Moerman
|
$\mathcal{M}$-points of bounded height
|
47 pages
| null | null | null |
math.NT math.AG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We initiate a general quantitative study of sets of $\mathcal{M}$-points, which are special subsets of rational points, generalizing Campana points, Darmon points, and squarefree solutions of Diophantine equations. We propose an asymptotic formula for the number of $\mathcal{M}$-points of bounded height on rationally connected varieties, extending Manin's conjecture as well as its generalization to Campana points by Pieropan, Smeets, Tanimoto and V\'arilly-Alvarado. Finally, we show that the conjecture explains several previously established results in arithmetic statistics.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 15:47:39 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 15:07:42 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Moerman",
"Boaz",
""
]
] |
2512.07661
|
Shihao Li
|
Shihao Li, Naisheng Ye, Tianyu Li, Kashyap Chitta, Tuo An, Peng Su, Boyang Wang, Haiou Liu, Chen Lv, Hongyang Li
|
Optimization-Guided Diffusion for Interactive Scene Generation
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Realistic and diverse multi-agent driving scenes are crucial for evaluating autonomous vehicles, but safety-critical events which are essential for this task are rare and underrepresented in driving datasets. Data-driven scene generation offers a low-cost alternative by synthesizing complex traffic behaviors from existing driving logs. However, existing models often lack controllability or yield samples that violate physical or social constraints, limiting their usability. We present OMEGA, an optimization-guided, training-free framework that enforces structural consistency and interaction awareness during diffusion-based sampling from a scene generation model. OMEGA re-anchors each reverse diffusion step via constrained optimization, steering the generation towards physically plausible and behaviorally coherent trajectories. Building on this framework, we formulate ego-attacker interactions as a game-theoretic optimization in the distribution space, approximating Nash equilibria to generate realistic, safety-critical adversarial scenarios. Experiments on nuPlan and Waymo show that OMEGA improves generation realism, consistency, and controllability, increasing the ratio of physically and behaviorally valid scenes from 32.35% to 72.27% for free exploration capabilities, and from 11% to 80% for controllability-focused generation. Our approach can also generate $5\times$ more near-collision frames with a time-to-collision under three seconds while maintaining the overall scene realism.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 15:56:18 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 15:08:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Li",
"Shihao",
""
],
[
"Ye",
"Naisheng",
""
],
[
"Li",
"Tianyu",
""
],
[
"Chitta",
"Kashyap",
""
],
[
"An",
"Tuo",
""
],
[
"Su",
"Peng",
""
],
[
"Wang",
"Boyang",
""
],
[
"Liu",
"Haiou",
""
],
[
"Lv",
"Chen",
""
],
[
"Li",
"Hongyang",
""
]
] |
2512.07702
|
Sangha Park
|
Sangha Park, Eunji Kim, Yeongtak Oh, Jooyoung Choi, Sungroh Yoon
|
Guiding What Not to Generate: Automated Negative Prompting for Text-Image Alignment
|
WACV 2026
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Despite substantial progress in text-to-image generation, achieving precise text-image alignment remains challenging, particularly for prompts with rich compositional structure or imaginative elements. To address this, we introduce Negative Prompting for Image Correction (NPC), an automated pipeline that improves alignment by identifying and applying negative prompts that suppress unintended content. We begin by analyzing cross-attention patterns to explain why both targeted negatives-those directly tied to the prompt's alignment error-and untargeted negatives-tokens unrelated to the prompt but present in the generated image-can enhance alignment. To discover useful negatives, NPC generates candidate prompts using a verifier-captioner-proposer framework and ranks them with a salient text-space score, enabling effective selection without requiring additional image synthesis. On GenEval++ and Imagine-Bench, NPC outperforms strong baselines, achieving 0.571 vs. 0.371 on GenEval++ and the best overall performance on Imagine-Bench. By guiding what not to generate, NPC provides a principled, fully automated route to stronger text-image alignment in diffusion models. Code is released at https://github.com/wiarae/NPC.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 16:49:19 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 06:42:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Park",
"Sangha",
""
],
[
"Kim",
"Eunji",
""
],
[
"Oh",
"Yeongtak",
""
],
[
"Choi",
"Jooyoung",
""
],
[
"Yoon",
"Sungroh",
""
]
] |
2512.07730
|
Sangha Park
|
Sangha Park, Seungryong Yoo, Jisoo Mok, Sungroh Yoon
|
SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object Hallucination
|
WACV 2026
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Although Multimodal Large Language Models (MLLMs) have advanced substantially, they remain vulnerable to object hallucination caused by language priors and visual information loss. To address this, we propose SAVE (Sparse Autoencoder-Driven Visual Information Enhancement), a framework that mitigates hallucination by steering the model along Sparse Autoencoder (SAE) latent features. A binary object-presence question-answering probe identifies the SAE features most indicative of the model's visual information processing, referred to as visual understanding features. Steering the model along these identified features reinforces grounded visual understanding and effectively reduces hallucination. With its simple design, SAVE outperforms state-of-the-art training-free methods on standard benchmarks, achieving a 10\%p improvement in CHAIR\_S and consistent gains on POPE and MMHal-Bench. Extensive evaluations across multiple models and layers confirm the robustness and generalizability of our approach. Further analysis reveals that steering along visual understanding features suppresses the generation of uncertain object tokens and increases attention to image tokens, mitigating hallucination. Code is released at https://github.com/wiarae/SAVE.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 17:20:07 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 06:37:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Park",
"Sangha",
""
],
[
"Yoo",
"Seungryong",
""
],
[
"Mok",
"Jisoo",
""
],
[
"Yoon",
"Sungroh",
""
]
] |
2512.07774
|
Andrew Kalish
|
Andrew Kalish, Pedro Fittipaldi de Castro, and Wladimir A. Benalcazar
|
A dynamical order parameter for the transition to nonergodic dynamics in the discrete nonlinear Schr\"odinger equation
| null | null | null | null |
cond-mat.stat-mech
|
http://creativecommons.org/licenses/by/4.0/
|
The discrete nonlinear Schr\"odinger equation (DNLSE) exhibits a transition from ergodic, delocalized dynamics to a weakly nonergodic regime characterized by breather formation; yet, a precise characterization of this transition has remained elusive. By sampling many microcanonically equivalent initial conditions, we identify the asymptotic ensemble variance of the Kolmogorov-Sinai entropy as a dynamical order parameter that vanishes in the ergodic phase and becomes finite once ergodicity is broken. The relaxation time governing the ensemble convergence of the KS entropy displays an essential singularity at the transition, yielding a sharp boundary between the two dynamical regimes. This framework provides a trajectory-independent method for detecting ergodicity breaking that is broadly applicable to nonlinear lattice systems with conserved quantities.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 17:56:42 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:36:03 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kalish",
"Andrew",
""
],
[
"de Castro",
"Pedro Fittipaldi",
""
],
[
"Benalcazar",
"Wladimir A.",
""
]
] |
2512.07899
|
Yunyan Yang
|
Juan Zhao, Jicheng Ma, Yunyan Yang, Liang Zhao
|
Finding core subgraphs of directed graphs via discrete Ricci curvature flow
|
21 pages
| null | null | null |
cs.SI math.AP math.CO
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Ricci curvature and its associated flow offer powerful geometric methods for analyzing complex networks. While existing research heavily focuses on applications for undirected graphs such as community detection and core extraction, there have been relatively less attention on directed graphs.
In this paper, we introduce a definition of Ricci curvature and an accompanying curvature flow for directed graphs. Crucially, for strongly connected directed graphs, this flow admits a unique global solution. We then apply this flow to detect strongly connected subgraphs from weakly connected directed graphs. (A weakly connected graph is connected overall but not necessarily strongly connected). Unlike prior work requiring graphs to be strongly connected, our method loosens this requirement. We transform a weakly connected graph into a strongly connected one by adding edges with very large artificial weights. This modification does not compromise our core subgraph detection. Due to their extreme weight, these added edges are automatically discarded during the final iteration of the Ricci curvature flow.
For core evaluation, our approach consistently surpasses traditional methods, achieving better results on at least two out of three key metrics. The implementation code is publicly available at https://github.com/12tangze12/Finding-core-subgraphs-on-directed-graphs.
|
[
{
"version": "v1",
"created": "Fri, 05 Dec 2025 15:36:41 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhao",
"Juan",
""
],
[
"Ma",
"Jicheng",
""
],
[
"Yang",
"Yunyan",
""
],
[
"Zhao",
"Liang",
""
]
] |
2512.07999
|
Robin {\O}stern Lien
|
Espen Robstad Jakobsen, Robin {\O}stern Lien, Artur Rutkowski
|
On Schauder estimates for nonlocal and mixed local-nonlocal viscous Hamilton$\unicode{x2013}$Jacobi equations
|
29 pages
| null | null | null |
math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove spatial Schauder estimates $\unicode{x2013}$ optimal regularity estimates in H\"older spaces $\unicode{x2013}$ and well-posedness results for mild and classical solutions of viscous Hamilton$\unicode{x2013}$Jacobi equations with subcritical nonlocal and mixed local-nonlocal diffusions in $\mathbb{R}^d$. Our results hold under mild assumptions on the nonlocal/mixed operators and Hamiltonians. The Laplacian, fractional Laplacians, nonsymmetric, spectrally one-sided, and strongly anisotropic integral operators, as well as sums of such operators are covered. We observe an interplay between the regularity of the initial data and the growth of the Hamiltonian in the gradient, and give results for the two canonical cases: (i) Lipschitz initial data and general Hamiltonians that are H\"older in space and merely locally Lipschitz in the gradient, and (ii) H\"older initial data and Hamiltonians that are H\"older in space and locally Lipschitz with power growth in the gradient. We compute explicit blow-up rates for $C^1$ and higher order H\"older norms as $t\to 0$. The results include short and long time existence of mild solutions, optimal regularity in H\"older spaces and corresponding Schauder a priori estimates, and that spatially smooth mild solutions are regular in time and pointwise classical solutions.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 19:39:01 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:29:12 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jakobsen",
"Espen Robstad",
""
],
[
"Lien",
"Robin Østern",
""
],
[
"Rutkowski",
"Artur",
""
]
] |
2512.08027
|
Charles Swanson
|
C.P.S. Swanson, S.T.A. Kumar, D.W. Dudt, E.R. Flom, W.B. Kalb, T.G. Kruger, M.F. Martin, J.R. Olatunji, S. Pasmann, L.Z. Tang, J. von der Linden, J. Wasserman, M. Avida, A.S. Basurto, M. Dickerson, N. de Boer, M.J. Donovan, A.H. Doudna Cate, D. Fort, W. Harris, U. Khera, A. Koen, J.A. Labbate, N. Maitra, A. Ottaviano, R.K. Parmar, E.J. Paul, B. Reydel, A. van Riel, P.K. Romano, M. Savastianov, S. Saxena, S. Seethalla, S. Srinivasan, R.H. Wu, D. Nash, J. Priebe, M. Slepchenkov, S. Walsh, B. Berzin, D.A. Gates, and the Thea Energy team
|
Overview of the Helios Design: A Practical Planar Coil Stellarator Fusion Power Plant
| null | null | null | null |
physics.plasm-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Thea Energy, Inc. has developed the preconceptual design for "Helios," a fusion power plant based on the planar coil stellarator architecture. In this overview paper, the design is summarized and the reader is referred to the other papers for more detail. The Helios design is based around a two-field-period quasi-axisymmetric ("QA") stellarator equilibrium with aspect ratio 4.5 and a novel tokamak-like X-point divertor. The natural stability, low recirculating power, and steady-state capability of the stellarator are leveraged. Stability and transport are calculated using state-of-the-art, high-fidelity codes and grounded in measured performance of existing experiments. The electromagnetic coil set is high-temperature superconducting ("HTS") and consists of 12 large, plasma-encircling coils like the toroidal field coils of a tokamak, and 324 smaller, field-shaping coils. All coils are planar and convex. A maximum of 20 T on-coil is enforced, a value which has been achieved in existing large-bore HTS coils. There is a minimum of 1.2 m between plasma and coils, leaving space for tritium breeding blanket and neutron shielding. Because of this thick shielding, all coils have a minimum 40-year operational lifetime, the same minimum lifetime of the power plant system. 1.1 GW of thermal power and 390 MW of net electric power are produced. The shaping coils are individually controllable, enabling a uniquely configurable magnetic field for relaxed manufacturing and assembly tolerances and plasma control. A practical maintenance architecture is a primary driver of the design; maintenance is performed on entire toroidal sectors that are removed from between the encircling coils. A biennial maintenance cycle is estimated to take approximately 84 days, resulting in an 88% capacity factor. Rigorous engineering constraints such as temperature and stress limits are enforced.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 20:38:49 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:24:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Swanson",
"C. P. S.",
""
],
[
"Kumar",
"S. T. A.",
""
],
[
"Dudt",
"D. W.",
""
],
[
"Flom",
"E. R.",
""
],
[
"Kalb",
"W. B.",
""
],
[
"Kruger",
"T. G.",
""
],
[
"Martin",
"M. F.",
""
],
[
"Olatunji",
"J. R.",
""
],
[
"Pasmann",
"S.",
""
],
[
"Tang",
"L. Z.",
""
],
[
"von der Linden",
"J.",
""
],
[
"Wasserman",
"J.",
""
],
[
"Avida",
"M.",
""
],
[
"Basurto",
"A. S.",
""
],
[
"Dickerson",
"M.",
""
],
[
"de Boer",
"N.",
""
],
[
"Donovan",
"M. J.",
""
],
[
"Cate",
"A. H. Doudna",
""
],
[
"Fort",
"D.",
""
],
[
"Harris",
"W.",
""
],
[
"Khera",
"U.",
""
],
[
"Koen",
"A.",
""
],
[
"Labbate",
"J. A.",
""
],
[
"Maitra",
"N.",
""
],
[
"Ottaviano",
"A.",
""
],
[
"Parmar",
"R. K.",
""
],
[
"Paul",
"E. J.",
""
],
[
"Reydel",
"B.",
""
],
[
"van Riel",
"A.",
""
],
[
"Romano",
"P. K.",
""
],
[
"Savastianov",
"M.",
""
],
[
"Saxena",
"S.",
""
],
[
"Seethalla",
"S.",
""
],
[
"Srinivasan",
"S.",
""
],
[
"Wu",
"R. H.",
""
],
[
"Nash",
"D.",
""
],
[
"Priebe",
"J.",
""
],
[
"Slepchenkov",
"M.",
""
],
[
"Walsh",
"S.",
""
],
[
"Berzin",
"B.",
""
],
[
"Gates",
"D. A.",
""
],
[
"team",
"the Thea Energy",
""
]
] |
2512.08052
|
Pedro Santana
|
Pedro Santana
|
An Introduction to Deep Reinforcement and Imitation Learning
| null | null | null | null |
cs.RO cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Embodied agents, such as robots and virtual characters, must continuously select actions to execute tasks effectively, solving complex sequential decision-making problems. Given the difficulty of designing such controllers manually, learning-based approaches have emerged as promising alternatives, most notably Deep Reinforcement Learning (DRL) and Deep Imitation Learning (DIL). DRL leverages reward signals to optimize behavior, while DIL uses expert demonstrations to guide learning. This document introduces DRL and DIL in the context of embodied agents, adopting a concise, depth-first approach to the literature. It is self-contained, presenting all necessary mathematical and machine learning concepts as they are needed. It is not intended as a survey of the field; rather, it focuses on a small set of foundational algorithms and techniques, prioritizing in-depth understanding over broad coverage. The material ranges from Markov Decision Processes to REINFORCE and Proximal Policy Optimization (PPO) for DRL, and from Behavioral Cloning to Dataset Aggregation (DAgger) and Generative Adversarial Imitation Learning (GAIL) for DIL.
|
[
{
"version": "v1",
"created": "Mon, 08 Dec 2025 21:21:01 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 00:05:29 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Santana",
"Pedro",
""
]
] |
2512.08170
|
Haoxin Zhang
|
Haoxin Zhang, Shuaixin Li, Xiaozhou Zhu, Hongbo Chen, Wen Yao
|
RAVES-Calib: Robust, Accurate and Versatile Extrinsic Self Calibration Using Optimal Geometric Features
| null | null | null | null |
cs.RO cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we present a user-friendly LiDAR-camera calibration toolkit that is compatible with various LiDAR and camera sensors and requires only a single pair of laser points and a camera image in targetless environments. Our approach eliminates the need for an initial transform and remains robust even with large positional and rotational LiDAR-camera extrinsic parameters. We employ the Gluestick pipeline to establish 2D-3D point and line feature correspondences for a robust and automatic initial guess. To enhance accuracy, we quantitatively analyze the impact of feature distribution on calibration results and adaptively weight the cost of each feature based on these metrics. As a result, extrinsic parameters are optimized by filtering out the adverse effects of inferior features. We validated our method through extensive experiments across various LiDAR-camera sensors in both indoor and outdoor settings. The results demonstrate that our method provides superior robustness and accuracy compared to SOTA techniques. Our code is open-sourced on GitHub to benefit the community.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 01:58:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Haoxin",
""
],
[
"Li",
"Shuaixin",
""
],
[
"Zhu",
"Xiaozhou",
""
],
[
"Chen",
"Hongbo",
""
],
[
"Yao",
"Wen",
""
]
] |
2512.08225
|
Justin Kaidi
|
Jan Albert, Yamato Honda, Justin Kaidi, Yunqin Zheng
|
Haagerup Symmetry in $(E_8)_1$?
|
13 pages, 4 figures
| null | null | null |
hep-th cond-mat.str-el math-ph math.CT math.MP math.QA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We suggest that the chiral $(\mathfrak{e}_8)_1$ theory -- in many senses the simplest VOA -- may have Haagerup symmetry $\mathcal{H}_i$ for $i=1,2,3$. Likewise, we suggest that the non-chiral $(E_8)_1$ WZW model may have $\mathcal{H}_i \times \mathcal{H}_i^\textrm{op}$ symmetry, and that gauging the diagonal symmetry gives a $c=8$ theory with $\mathcal{Z}(\mathcal{H}_3)$ symmetry, which is the theory predicted in \cite{Evans:2010yr}. Along the way, we show that $(E_8)_1$ also has a $\mathrm{Fib} \times \mathrm{Fib}^\text{op}$ symmetry, and that gauging the diagonal symmetry gives the $(G_2)_1 \times (F_4)_1$ WZW model, explaining the well-known conformal embedding $(G_2)_1 \times (F_4)_1 \subset (E_8)_1$. Finally, we suggest a relation to theories with $\mathcal{H}_3$ symmetry at $c=2,6$, complimenting the discussion with new modular bootstrap results.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 04:05:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Albert",
"Jan",
""
],
[
"Honda",
"Yamato",
""
],
[
"Kaidi",
"Justin",
""
],
[
"Zheng",
"Yunqin",
""
]
] |
2512.08260
|
Qing-Zeng Yan
|
Ji Yang, Qing-Zeng Yan, Yang Su, Shaobo Zhang, Xin Zhou, Yan Sun, Yiping Ao, Xuepeng Chen, Zhiwei Chen, Fujun Du, Min Fang, Yan Gong, Zhibo Jiang, Shengyu Jin, Binggang Ju, Chong Li, Yingjie Li, Yi Liu, Dengrong Lu, Chunsheng Luo, Yuehui Ma, Ruiqing Mao, Jixian Sun, Chen Wang, Hongchi Wang, Min Wang, Min Wang (Qinghai), Xindong Wang, Wenting Xu, Ye Xu, Kun Yan, Ping Yan, Lixia Yuan, Miaomiao Zhang, Yongxing Zhang
|
The Milky Way Imaging Scroll Painting Survey: Data Release 1
|
Accepted for publication in ApJS. MWISP DR1 link: https://doi.org/10.57760/sciencedb.27351
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We present the first data release (DR1) of the Milky Way Imaging Scroll Painting (MWISP) survey, a mapping in the J=(1-0) transition lines of 12CO, 13CO, and C18O toward the northern Galactic plane during 2011-2022. The MWISP survey was conducted using the PMO 13.7 m telescope at a spatial resolution of approximately 50" and a velocity resolution of 0.16 km/s at 115 GHz. DR1 fully covered 2310 square degrees within the Galactic longitude (l) and latitude (b) range of 9.75 deg =< l =< 229.75 deg and |b| =< 5.25 deg. The surveyed area was divided into cell units of 30'x30' for practical purposes and On-The-Fly (OTF) mapping was performed toward each target cell unit. The data were regridded into a regular 3D datacube in l-b-V_LSR with a pixel size of 30" in l-b axes and 0.16 km/s in theV_LSR axis. The median rms noise is 0.47 K, 0.25 K, and 0.25 K for 12CO, 13CO, and C18O, respectively. The equivalent 3 sigma sensitivity in 12CO luminosity is approximately 0.23 K km/s, making MWISP the most sensitive survey of its kind. In this paper, we describe the survey data, including the calibration, data cleaning, data mosaic processes, and the data products. The final mosaicked data cubes contain about 3.33x10^7 spectra (pixels) for each CO isotopologue line. Color composite images, made from the intensities of the isotopologue lines, and some concise descriptions are provided. We constructed a molecular cloud catalog based on the mosaicked 12CO data cube using the clustering algorithm DBSCAN, detecting 103,517 molecular clouds, 10,790 of which exhibit 13CO emission and 304 of which show C18O emission. Based on the histogram of voxel brightness temperature, we estimated a total 12CO flux of 7.69+/-0.38x10^7 K km/s arcmin^2, 82% of which is captured by the DBSCAN algorithm. The data, together with the cloud sample, provide unique information on molecular gas in the northern Milky Way.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 05:37:48 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 01:17:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yang",
"Ji",
"",
"Qinghai"
],
[
"Yan",
"Qing-Zeng",
"",
"Qinghai"
],
[
"Su",
"Yang",
"",
"Qinghai"
],
[
"Zhang",
"Shaobo",
"",
"Qinghai"
],
[
"Zhou",
"Xin",
"",
"Qinghai"
],
[
"Sun",
"Yan",
"",
"Qinghai"
],
[
"Ao",
"Yiping",
"",
"Qinghai"
],
[
"Chen",
"Xuepeng",
"",
"Qinghai"
],
[
"Chen",
"Zhiwei",
"",
"Qinghai"
],
[
"Du",
"Fujun",
"",
"Qinghai"
],
[
"Fang",
"Min",
"",
"Qinghai"
],
[
"Gong",
"Yan",
"",
"Qinghai"
],
[
"Jiang",
"Zhibo",
"",
"Qinghai"
],
[
"Jin",
"Shengyu",
"",
"Qinghai"
],
[
"Ju",
"Binggang",
"",
"Qinghai"
],
[
"Li",
"Chong",
"",
"Qinghai"
],
[
"Li",
"Yingjie",
"",
"Qinghai"
],
[
"Liu",
"Yi",
"",
"Qinghai"
],
[
"Lu",
"Dengrong",
"",
"Qinghai"
],
[
"Luo",
"Chunsheng",
"",
"Qinghai"
],
[
"Ma",
"Yuehui",
"",
"Qinghai"
],
[
"Mao",
"Ruiqing",
"",
"Qinghai"
],
[
"Sun",
"Jixian",
"",
"Qinghai"
],
[
"Wang",
"Chen",
"",
"Qinghai"
],
[
"Wang",
"Hongchi",
"",
"Qinghai"
],
[
"Wang",
"Min",
"",
"Qinghai"
],
[
"Wang",
"Min",
"",
"Qinghai"
],
[
"Wang",
"Xindong",
""
],
[
"Xu",
"Wenting",
""
],
[
"Xu",
"Ye",
""
],
[
"Yan",
"Kun",
""
],
[
"Yan",
"Ping",
""
],
[
"Yuan",
"Lixia",
""
],
[
"Zhang",
"Miaomiao",
""
],
[
"Zhang",
"Yongxing",
""
]
] |
2512.08390
|
Jean-Philip Piquemal
|
Daniele Loco, Kisa Barkemeyer, Andre R. R. Carvalho, Jean-Philip Piquemal
|
Practical protein-pocket hydration-site prediction for drug discovery on a quantum computer
| null | null | null | null |
quant-ph physics.bio-ph physics.chem-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Demonstrating the practical utility of Noisy Intermediate-Scale Quantum (NISQ) hardware for recurrent tasks in Computer-Aided Drug Discovery is of paramount importance. We tackle this challenge by performing three-dimensional protein pockets hydration-site prediction on a quantum computer. Formulating the water placement problem as a Quadratic Unconstrained Binary Optimization (QUBO), we use a hybrid approach coupling a classical three-dimensional reference-interaction site model (3D-RISM) to an efficient quantum optimization solver, to run various hardware experiments up to 123 qubits. Matching the precision of classical approaches, our results reproduced experimental predictions on real-life protein-ligand complexes. Furthermore, through a detailed resource estimation analysis, we show that accuracy can be systematically improved with increasing number of qubits, indicating that full quantum utility is in reach. Finally, we provide evidence that advantageous situations could be found for systems where classical optimization struggles to provide optimal solutions. The method has potential for assisting simulations of protein-ligand complexes for drug lead optimization and setup of docking calculations.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 09:20:15 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:53:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Loco",
"Daniele",
""
],
[
"Barkemeyer",
"Kisa",
""
],
[
"Carvalho",
"Andre R. R.",
""
],
[
"Piquemal",
"Jean-Philip",
""
]
] |
2512.08400
|
Samitha Nuwan Thilakarathna
|
Samitha Nuwan Thilakarathna, Ercan Avsar, Martin Mathias Nielsen, Malte Pedersen
|
Towards Visual Re-Identification of Fish using Fine-Grained Classification for Electronic Monitoring in Fisheries
|
The paper has been accepted for publication at Northern Lights Deep Learning (NLDL) Conference 2026
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Accurate fisheries data are crucial for effective and sustainable marine resource management. With the recent adoption of Electronic Monitoring (EM) systems, more video data is now being collected than can be feasibly reviewed manually. This paper addresses this challenge by developing an optimized deep learning pipeline for automated fish re-identification (Re-ID) using the novel AutoFish dataset, which simulates EM systems with conveyor belts with six similarly looking fish species. We demonstrate that key Re-ID metrics (R1 and mAP@k) are substantially improved by using hard triplet mining in conjunction with a custom image transformation pipeline that includes dataset-specific normalization. By employing these strategies, we demonstrate that the Vision Transformer-based Swin-T architecture consistently outperforms the Convolutional Neural Network-based ResNet-50, achieving peak performance of 41.65% mAP@k and 90.43% Rank-1 accuracy. An in-depth analysis reveals that the primary challenge is distinguishing visually similar individuals of the same species (Intra-species errors), where viewpoint inconsistency proves significantly more detrimental than partial occlusion. The source code and documentation are available at: https://github.com/msamdk/Fish_Re_Identification.git
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 09:33:53 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 11:39:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Thilakarathna",
"Samitha Nuwan",
""
],
[
"Avsar",
"Ercan",
""
],
[
"Nielsen",
"Martin Mathias",
""
],
[
"Pedersen",
"Malte",
""
]
] |
2512.08417
|
Yinan Zhong
|
Yinan Zhong, Qianhao Miao, Yanjiao Chen, Jiangyi Deng, Yushi Cheng, Wenyuan Xu
|
Attention is All You Need to Defend Against Indirect Prompt Injection Attacks in LLMs
|
Accepted by Network and Distributed System Security (NDSS) Symposium 2026
| null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large Language Models (LLMs) have been integrated into many applications (e.g., web agents) to perform more sophisticated tasks. However, LLM-empowered applications are vulnerable to Indirect Prompt Injection (IPI) attacks, where instructions are injected via untrustworthy external data sources. This paper presents Rennervate, a defense framework to detect and prevent IPI attacks. Rennervate leverages attention features to detect the covert injection at a fine-grained token level, enabling precise sanitization that neutralizes IPI attacks while maintaining LLM functionalities. Specifically, the token-level detector is materialized with a 2-step attentive pooling mechanism, which aggregates attention heads and response tokens for IPI detection and sanitization. Moreover, we establish a fine-grained IPI dataset, FIPI, to be open-sourced to support further research. Extensive experiments verify that Rennervate outperforms 15 commercial and academic IPI defense methods, achieving high precision on 5 LLMs and 6 datasets. We also demonstrate that Rennervate is transferable to unseen attacks and robust against adaptive adversaries.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 09:44:13 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:47:12 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhong",
"Yinan",
""
],
[
"Miao",
"Qianhao",
""
],
[
"Chen",
"Yanjiao",
""
],
[
"Deng",
"Jiangyi",
""
],
[
"Cheng",
"Yushi",
""
],
[
"Xu",
"Wenyuan",
""
]
] |
2512.08442
|
Aleksandr Dyatlov
|
A.S. Dyatlov, D.M. Dolgintsev, V.V. Gerasimov, V.V. Kobets, V.P. Nazmov, M.A. Nozdrin, A.N. Sergeev, D.S. Shokin, K.E. Yunenko, and D.V. Karlovets
|
High-OAM Deep Ultraviolet Twisted Light Generation for RF-Photoinjector Applications
| null | null | null | null |
quant-ph physics.acc-ph physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
We report on the generation and characterization of ultraviolet (wavelength 266 nm) twisted light with high orbital angular momentum (OAM) using three types of fabricated diffractive optical elements (DOEs): a reflective fork grating, a high-charge spiral phase plate (SPP), and binary axicons. All elements were integrated into a drive-laser beamline of an electron RF-photoinjector, enabling direct evaluation under accelerator-relevant conditions. The SPP produced a high-purity Laguerre-Gaussian mode with OAM l = 64 and a measured conversion efficiency of approximately 80%. Binary axicons generated quasi-Bessel twisted light with topological charges up to m = 10, exhibiting low divergence and stable multi-lobe ring structures. The fork grating reliably produced lower-order modes, l = 2-8, with good agreement between simulations and cylindrical-lens diagnostics. These results constitute, to our knowledge, the first comprehensive experimental demonstration of deep-UV high-OAM beams generated with fabricated DOEs and validated through mode-conversion measurements. The demonstrated techniques are compatible with high-power UV laser systems used in RF-photoinjectors and offer a practical route toward structured photocathode illumination and the generation of relativistic vortex electrons at a particle accelerator facility.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 10:15:02 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 09:07:19 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 08:45:42 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dyatlov",
"A. S.",
""
],
[
"Dolgintsev",
"D. M.",
""
],
[
"Gerasimov",
"V. V.",
""
],
[
"Kobets",
"V. V.",
""
],
[
"Nazmov",
"V. P.",
""
],
[
"Nozdrin",
"M. A.",
""
],
[
"Sergeev",
"A. N.",
""
],
[
"Shokin",
"D. S.",
""
],
[
"Yunenko",
"K. E.",
""
],
[
"Karlovets",
"D. V.",
""
]
] |
2512.08470
|
Ksenia Shagalov
|
Ksenia Shagalov, David Feldstein-Bofill, Leo Uhre Jakobsen, Zhenhai Sun, Casper Wied, Amalie T. J. Paulsen, Johann Bock Severin, Malthe A. Marciniak, Clinton A. Potts, Anders Kringh{\o}j, Jacob Hastrup, Karsten Flensberg, Svend Kr{\o}jer, Morten Kjaergaard
|
Higher Josephson harmonics in a tunable double-junction transmon qubit
| null | null | null | null |
quant-ph cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
Tunable Josephson harmonics open new avenues for qubit design. We demonstrate a superconducting circuit element consisting of a tunnel junction in series with a SQUID loop, yielding a Josephson potential whose harmonic content is strongly tunable by magnetic flux. Through spectroscopy of the first four qubit transitions, together with an effective single-mode model renormalized by the internal mode, we resolve a second harmonic with an amplitude up to $\sim10\%$ of the fundamental. We identify a flux sweet spot where the dispersive shift vanishes, achieved by balancing the dispersive couplings to the internal and qubit modes. This highly tunable element provides a route toward protected qubits and customizable nonlinear microwave devices.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 10:42:06 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:24:14 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shagalov",
"Ksenia",
""
],
[
"Feldstein-Bofill",
"David",
""
],
[
"Jakobsen",
"Leo Uhre",
""
],
[
"Sun",
"Zhenhai",
""
],
[
"Wied",
"Casper",
""
],
[
"Paulsen",
"Amalie T. J.",
""
],
[
"Severin",
"Johann Bock",
""
],
[
"Marciniak",
"Malthe A.",
""
],
[
"Potts",
"Clinton A.",
""
],
[
"Kringhøj",
"Anders",
""
],
[
"Hastrup",
"Jacob",
""
],
[
"Flensberg",
"Karsten",
""
],
[
"Krøjer",
"Svend",
""
],
[
"Kjaergaard",
"Morten",
""
]
] |
2512.08511
|
Wenxi Yang
|
Wenxi Yang, Yuzhong Zhao, Fang Wan, Qixiang Ye
|
Thinking with Images via Self-Calling Agent
|
Code is available at https://github.com/YWenxi/think-with-images-through-self-calling
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Thinking-with-images paradigms have showcased remarkable visual reasoning capability by integrating visual information as dynamic elements into the Chain-of-Thought (CoT). However, optimizing interleaved multimodal CoT (iMCoT) through reinforcement learning remains challenging, as it relies on scarce high-quality reasoning data. In this study, we propose Self-Calling Chain-of-Thought (sCoT), a novel visual reasoning paradigm that reformulates iMCoT as a language-only CoT with self-calling. Specifically, a main agent decomposes the complex visual reasoning task to atomic subtasks and invokes its virtual replicas, i.e. parameter-sharing subagents, to solve them in isolated context. sCoT enjoys substantial training effectiveness and efficiency, as it requires no explicit interleaving between modalities. sCoT employs group-relative policy optimization to reinforce effective reasoning behavior to enhance optimization. Experiments on HR-Bench 4K show that sCoT improves the overall reasoning performance by up to $1.9\%$ with $\sim 75\%$ fewer GPU hours compared to strong baseline approaches. Code is available at https://github.com/YWenxi/think-with-images-through-self-calling.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 11:53:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 13:21:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yang",
"Wenxi",
""
],
[
"Zhao",
"Yuzhong",
""
],
[
"Wan",
"Fang",
""
],
[
"Ye",
"Qixiang",
""
]
] |
2512.08515
|
Zhi-Long Han
|
Zhen-Wei Wang, Zhi-Long Han, Fei Huang, Honglei Li, Ang Liu
|
Reviving $Z^\prime$ Portal Dark Matter with Conversion Mechanism
|
32 pages, 10 figures
| null | null | null |
hep-ph hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
In many new physics models with extended gauge symmetry, the new gauge boson $Z'$ could mediate the interactions between the dark matter and standard model particles. For the conventional $Z^\prime$ portal dark matter, the collider and the direct detection constraints typically pose a significant challenge. To address this pressing issue, we present in this paper a new benchmark model based on the gauged $U(1)_{B-L}$ symmetry, which introduces a Dirac dark fermion $\tilde{\chi}_1$ and a heavier partner $\tilde{\chi}_2$ with zero and nonzero $U(1)_{B-L}$ charge, respectively. Including the mass term $\delta m \bar{\tilde{\chi}}_1\tilde{\chi}_2$ results in the dark fermions $\chi_1$ and $\chi_2$ in the mass eigenstate, where the lighter one $\chi_1$ is regarded as the dark matter candidate. Various intriguing processes for the relic density arise with the compressed mass spectrum $m_{\chi_1}\simeq m_{\chi_2}$, such as the coscattering $\chi_2f\to\chi_1f$, the conversion $\chi_2\chi_i\to\chi_1\chi_j$, and the coannihilation $\chi_1\chi_2\to f\bar{f}$ processes. Suppressed by the small mixing angle $\theta$ between the dark fermions, the small effective gauge coupling of dark matter $\chi_1$ to the gauge boson $Z'$ is one distinct feature of this model, rendering phenomenology in many aspects more promising. In this paper, we investigate the production of dark matter through new mechanisms within the frameworks of resonance and secluded scenarios. The impacts of phenomenological constraints from collider, dark matter, and cosmology are also taken into account. We report that the conversion mechanism is both favored by the resonance and secluded scenarios under current constraints.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 12:01:43 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:53:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Zhen-Wei",
""
],
[
"Han",
"Zhi-Long",
""
],
[
"Huang",
"Fei",
""
],
[
"Li",
"Honglei",
""
],
[
"Liu",
"Ang",
""
]
] |
2512.08609
|
Yang Liu
|
Hui Wang, Yang Liu, Xiaoyu Zhang, Chaoxu Mu
|
CogMCTS: A Novel Cognitive-Guided Monte Carlo Tree Search Framework for Iterative Heuristic Evolution with Large Language Models
| null | null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Automatic Heuristic Design (AHD) is an effective framework for solving complex optimization problems. The development of large language models (LLMs) enables the automated generation of heuristics. Existing LLM-based evolutionary methods rely on population strategies and are prone to local optima. Integrating LLMs with Monte Carlo Tree Search (MCTS) improves the trade-off between exploration and exploitation, but multi-round cognitive integration remains limited and search diversity is constrained. To overcome these limitations, this paper proposes a novel cognitive-guided MCTS framework (CogMCTS). CogMCTS tightly integrates the cognitive guidance mechanism of LLMs with MCTS to achieve efficient automated heuristic optimization. The framework employs multi-round cognitive feedback to incorporate historical experience, node information, and negative outcomes, dynamically improving heuristic generation. Dual-track node expansion combined with elite heuristic management balances the exploration of diverse heuristics and the exploitation of high-quality experience. In addition, strategic mutation modifies the heuristic forms and parameters to further enhance the diversity of the solution and the overall optimization performance. The experimental results indicate that CogMCTS outperforms existing LLM-based AHD methods in stability, efficiency, and solution quality.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 13:54:18 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:46:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Hui",
""
],
[
"Liu",
"Yang",
""
],
[
"Zhang",
"Xiaoyu",
""
],
[
"Mu",
"Chaoxu",
""
]
] |
2512.08671
|
Huzaifa Arif
|
Huzaifa Arif
|
DS FedProxGrad: Asymptotic Stationarity Without Noise Floor in Fair Federated Learning
|
8 pages
| null | null | null |
cs.LG stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent work \cite{arifgroup} introduced Federated Proximal Gradient \textbf{(\texttt{FedProxGrad})} for solving non-convex composite optimization problems in group fair federated learning. However, the original analysis established convergence only to a \textit{noise-dominated neighborhood of stationarity}, with explicit dependence on a variance-induced noise floor. In this work, we provide an improved asymptotic convergence analysis for a generalized \texttt{FedProxGrad}-type analytical framework with inexact local proximal solutions and explicit fairness regularization. We call this extended analytical framework \textbf{DS \texttt{FedProxGrad}} (Decay Step Size \texttt{FedProxGrad}). Under a Robbins-Monro step-size schedule \cite{robbins1951stochastic} and a mild decay condition on local inexactness, we prove that $\liminf_{r\to\infty} \mathbb{E}[\|\nabla F(\mathbf{x}^r)\|^2] = 0$, i.e., the algorithm is asymptotically stationary and the convergence rate does not depend on a variance-induced noise floor.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 14:55:21 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 17:10:07 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 02:35:40 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Arif",
"Huzaifa",
""
]
] |
2512.08682
|
Walter Rocchia
|
Sergii V. Siryk, Walter Rocchia
|
Many interacting particles in solution. II. Screening-ranged expansion of electrostatic forces
| null | null | null | null |
cond-mat.soft math-ph math.MP physics.bio-ph physics.chem-ph physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a fully analytical integration of the Maxwell stress tensor and derive exact relations for interparticle forces in systems of multiple dielectric spheres immersed in a polarizable ionic solvent, within the framework of the linearized Poisson--Boltzmann theory. Building upon the screening-ranged (in ascending orders of Debye screening) expansions of the potentials developed and rigorously analyzed in the accompanying works arXiv:2512.08407, arXiv:2512.08684, arXiv:2512.09421, we construct exact screening-ranged many-body expansions for electrostatic forces in explicit analytical form. These results establish a rigorous foundation for evaluating screened electrostatic interactions in complex particle systems and provide direct analytical connections to, and systematic improvements upon, various earlier approximate or limited-case formulations available in the literature, both at zero and finite ionic strength.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 15:05:53 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:49:34 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Siryk",
"Sergii V.",
""
],
[
"Rocchia",
"Walter",
""
]
] |
2512.08684
|
Walter Rocchia
|
Sergii V. Siryk, Walter Rocchia
|
Many interacting particles in solution. III. Spectral analysis of the associated Neumann--Poincar\'e-type operators
| null | null | null | null |
cond-mat.soft math-ph math.MP physics.bio-ph physics.chem-ph physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The interaction of particles in an electrolytic medium can be calculated by solving the Poisson equation inside the solutes and the linearized Poisson--Boltzmann equation in the solvent, with suitable boundary conditions at the interfaces. Analytical approaches often expand the potentials in spherical harmonics, relating interior and exterior coefficients and eliminating some coefficients in favor of others, but a rigorous spectral analysis of the corresponding formulations is still lacking. Here, we introduce pertinent composite many-body Neumann--Poincar\'e-type operators and prove that they are compact with spectral radii strictly less than one. These results provide the foundation for systematic screening-ranged expansions, in powers of the Debye screening parameters, of electrostatic potentials, interaction energies, and forces, and establish the analytical framework for the accompanying works arXiv:2512.09421, arXiv:2512.08407, arXiv:2512.08682.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 15:06:19 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:52:47 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Siryk",
"Sergii V.",
""
],
[
"Rocchia",
"Walter",
""
]
] |
2512.08743
|
Haoyang Yan
|
Shuyue Hu, Haoyang Yan, Yiqun Zhang, Yang Chen, Dongzhan Zhou, Lei Bai
|
Towards Foundation Models with Native Multi-Agent Intelligence
| null | null | null | null |
cs.AI cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Foundation models (FMs) are increasingly assuming the role of the "brain" of AI agents. While recent efforts have begun to equip FMs with native single-agent abilities -- such as GUI interaction or integrated tool use -- we argue that the next frontier is endowing FMs with native multi-agent intelligence. We identify four core capabilities of FMs in multi-agent contexts: understanding, planning, efficient communication, and adaptation. Contrary to assumptions about the spontaneous emergence of such abilities, we provide extensive empirical evidence across 41 large language models showing that strong single-agent performance alone does not automatically yield robust multi-agent intelligence. To address this gap, we outline key research directions -- spanning dataset construction, evaluation, training paradigms, and safety considerations -- for building FMs with native multi-agent intelligence.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 15:51:36 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:06:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hu",
"Shuyue",
""
],
[
"Yan",
"Haoyang",
""
],
[
"Zhang",
"Yiqun",
""
],
[
"Chen",
"Yang",
""
],
[
"Zhou",
"Dongzhan",
""
],
[
"Bai",
"Lei",
""
]
] |
2512.08796
|
Anirban Mukhopadhyay
|
Anirban Mukhopadhyay, Ihor I. Syvorotka, and Anil Prabhakar
|
Injection dynamics in spin-wave active ring oscillator (SWARO)
|
12 pages, 5 figures
| null | null | null |
physics.app-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We investigated injection locking in spin-wave active ring oscillators (SWAROs) operating in the multi-mode regime. By applying external RF signals with varying frequency and power, we identified the locking behavior of individual modes and extracted the total locking ranges from spectral measurements. The results show asymmetric evolution of the lower and upper locking boundaries with drive power for the lower-frequency SWARO modes, while the highest-frequency mode exhibits nearly symmetric behavior. A maximum locking range of over 11 MHz is observed at a drive power of -10 dBm. To interpret these results, we develop an Adler-like model that captures the dependence of the locking range on drive power, showing good agreement for the higher-frequency modes. For the lowest-frequency mode, however, the model underestimates the locking range at low drive and saturates at high drive power levels, while the experimental range increases monotonically, indicating the influence of multi-mode interactions. These findings establish SWARO as a useful platform for exploring injection phenomena in spin-wave ring systems with delayed feedback and motivate the development of extended injection models that account for multi-mode dynamics.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 16:48:20 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:13:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mukhopadhyay",
"Anirban",
""
],
[
"Syvorotka",
"Ihor I.",
""
],
[
"Prabhakar",
"Anil",
""
]
] |
2512.08844
|
Stephen Barrett Dr
|
Malcolm Murray, Steve Barrett, Henry Papadatos, Otter Quarks, Matt Smith, Alejandro Tlaie Boria, Chlo\'e Touzet, Sim\'eon Campos
|
A Methodology for Quantitative AI Risk Modeling
|
The only changes in v2 are some updates to a few arXiv URLs in the references
| null | null | null |
cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
Although general-purpose AI systems offer transformational opportunities in science and industry, they simultaneously raise critical concerns about safety, misuse, and potential loss of control. Despite these risks, methods for assessing and managing them remain underdeveloped. Effective risk management requires systematic modeling to characterize potential harms, as emphasized in frameworks such as the EU General-Purpose AI Code of Practice. This paper advances the risk modeling component of AI risk management by introducing a methodology that integrates scenario building with quantitative risk estimation, drawing on established approaches from other high-risk industries. Our methodology models risks through a six-step process: (1) defining risk scenarios, (2) decomposing them into quantifiable parameters, (3) quantifying baseline risk without AI models, (4) identifying key risk indicators such as benchmarks, (5) mapping these indicators to model parameters to estimate LLM uplift, and (6) aggregating individual parameters into risk estimates that enable concrete claims (e.g., X% probability of >\$Y in annual cyber damages). We examine the choices that underlie our methodology throughout the article, with discussions of strengths, limitations, and implications for future research. Our methodology is designed to be applicable to key systemic AI risks, including cyber offense, biological weapon development, harmful manipulation, and loss-of-control, and is validated through extensive application in LLM-enabled cyber offense. Detailed empirical results and cyber-specific insights are presented in a companion paper.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 17:34:59 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:25:12 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Murray",
"Malcolm",
""
],
[
"Barrett",
"Steve",
""
],
[
"Papadatos",
"Henry",
""
],
[
"Quarks",
"Otter",
""
],
[
"Smith",
"Matt",
""
],
[
"Boria",
"Alejandro Tlaie",
""
],
[
"Touzet",
"Chloé",
""
],
[
"Campos",
"Siméon",
""
]
] |
2512.08864
|
Jakub Kry\'s
|
Steve Barrett, Malcolm Murray, Otter Quarks, Matthew Smith, Jakub Kry\'s, Sim\'eon Campos, Alejandro Tlaie Boria, Chlo\'e Touzet, Sevan Hayrapet, Fred Heiding, Omer Nevo, Adam Swanda, Jair Aguirre, Asher Brass Gershovich, Eric Clay, Ryan Fetterman, Mario Fritz, Marc Juarez, Vasilios Mavroudis, Henry Papadatos
|
Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse
| null | null | null | null |
cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
Advanced AI systems offer substantial benefits but also introduce risks. In 2025, AI-enabled cyber offense has emerged as a concrete example. This technical report applies a quantitative risk modeling methodology (described in full in a companion paper) to this domain. We develop nine detailed cyber risk models that allow analyzing AI uplift as a function of AI benchmark performance. Each model decomposes attacks into steps using the MITRE ATT&CK framework and estimates how AI affects the number of attackers, attack frequency, probability of success, and resulting harm to determine different types of uplift. To produce these estimates with associated uncertainty, we employ both human experts, via a Delphi study, as well as LLM-based simulated experts, both mapping benchmark scores (from Cybench and BountyBench) to risk model factors. Individual estimates are aggregated through Monte Carlo simulation. The results indicate systematic uplift in attack efficacy, speed, and target reach, with different mechanisms of uplift across risk models. We aim for our quantitative risk modeling to fulfill several aims: to help cybersecurity teams prioritize mitigations, AI evaluators design benchmarks, AI developers make more informed deployment decisions, and policymakers obtain information to set risk thresholds. Similar goals drove the shift from qualitative to quantitative assessment over time in other high-risk industries, such as nuclear power. We propose this methodology and initial application attempt as a step in that direction for AI risk management. While our estimates carry significant uncertainty, publishing detailed quantified results can enable experts to pinpoint exactly where they disagree. This helps to collectively refine estimates, something that cannot be done with qualitative assessments alone.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 17:54:17 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:54:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Barrett",
"Steve",
""
],
[
"Murray",
"Malcolm",
""
],
[
"Quarks",
"Otter",
""
],
[
"Smith",
"Matthew",
""
],
[
"Kryś",
"Jakub",
""
],
[
"Campos",
"Siméon",
""
],
[
"Boria",
"Alejandro Tlaie",
""
],
[
"Touzet",
"Chloé",
""
],
[
"Hayrapet",
"Sevan",
""
],
[
"Heiding",
"Fred",
""
],
[
"Nevo",
"Omer",
""
],
[
"Swanda",
"Adam",
""
],
[
"Aguirre",
"Jair",
""
],
[
"Gershovich",
"Asher Brass",
""
],
[
"Clay",
"Eric",
""
],
[
"Fetterman",
"Ryan",
""
],
[
"Fritz",
"Mario",
""
],
[
"Juarez",
"Marc",
""
],
[
"Mavroudis",
"Vasilios",
""
],
[
"Papadatos",
"Henry",
""
]
] |
2512.08868
|
Rui Min
|
Rui Min, Zile Qiao, Ze Xu, Jiawen Zhai, Wenyu Gao, Xuanzhong Chen, Haozhen Sun, Zhen Zhang, Xinyu Wang, Hong Zhou, Wenbiao Yin, Bo Zhang, Xuan Zhou, Ming Yan, Yong Jiang, Haicheng Liu, Liang Ding, Ling Zou, Yi R. Fung, Yalong Li, Pengjun Xie
|
EcomBench: Towards Holistic Evaluation of Foundation Agents in E-commerce
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Foundation agents have rapidly advanced in their ability to reason and interact with real environments, making the evaluation of their core capabilities increasingly important. While many benchmarks have been developed to assess agent performance, most concentrate on academic settings or artificially designed scenarios while overlooking the challenges that arise in real applications. To address this issue, we focus on a highly practical real-world setting, the e-commerce domain, which involves a large volume of diverse user interactions, dynamic market conditions, and tasks directly tied to real decision-making processes. To this end, we introduce EcomBench, a holistic E-commerce Benchmark designed to evaluate agent performance in realistic e-commerce environments. EcomBench is built from genuine user demands embedded in leading global e-commerce ecosystems and is carefully curated and annotated through human experts to ensure clarity, accuracy, and domain relevance. It covers multiple task categories within e-commerce scenarios and defines three difficulty levels that evaluate agents on key capabilities such as deep information retrieval, multi-step reasoning, and cross-source knowledge integration. By grounding evaluation in real e-commerce contexts, EcomBench provides a rigorous and dynamic testbed for measuring the practical capabilities of agents in modern e-commerce.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 18:00:26 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:38:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Min",
"Rui",
""
],
[
"Qiao",
"Zile",
""
],
[
"Xu",
"Ze",
""
],
[
"Zhai",
"Jiawen",
""
],
[
"Gao",
"Wenyu",
""
],
[
"Chen",
"Xuanzhong",
""
],
[
"Sun",
"Haozhen",
""
],
[
"Zhang",
"Zhen",
""
],
[
"Wang",
"Xinyu",
""
],
[
"Zhou",
"Hong",
""
],
[
"Yin",
"Wenbiao",
""
],
[
"Zhang",
"Bo",
""
],
[
"Zhou",
"Xuan",
""
],
[
"Yan",
"Ming",
""
],
[
"Jiang",
"Yong",
""
],
[
"Liu",
"Haicheng",
""
],
[
"Ding",
"Liang",
""
],
[
"Zou",
"Ling",
""
],
[
"Fung",
"Yi R.",
""
],
[
"Li",
"Yalong",
""
],
[
"Xie",
"Pengjun",
""
]
] |
2512.08898
|
Helena Ubach
|
Helena Ubach
|
Self-lensing of moving gravitational-wave sources can break the microlensing crossing timescale degeneracy
|
9 pages, 6 figures. Submitted version. Comments welcome
| null | null | null |
astro-ph.HE astro-ph.CO astro-ph.GA
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
When a moving gravitational-wave (GW) source travels behind a massive astrophysical object, its signal is gravitationally lensed, showing a waveform distortion similar to a Paczy\'{n}ski curve. We present a first study of the lensing signature of a massive black hole (MBH) on a frequency-dependent GW signal from a moving binary merger. For both light and GW sources in a Keplerian circular orbit around a MBH lens, the self-lensing geometry breaks the microlensing degeneracy in the Einstein radius crossing timescale $t_{\rm E}$. The duration of the curve ($2 t_{\rm E}$) becomes independent on the MBH mass $M_{\rm MBH}$, and provides a direct measure of the distance $d_{\rm LS}$ to the MBH. However, $M_{\rm MBH}$ remains unknown. We show that, in GW signals, the redshifted mass $M_{{\rm MBH},z}$ can additionally be obtained from the interference pattern, by measuring the modulation period $T$, the GW frequency $f$, and $t_{\rm E}$: $M_{{\rm MBH},z}\simeq 2.5\times 10^6\,M_\odot\,(t_{\rm E}/[100\,{\rm s}])\,(f\,T)^{-1}$. If this lensing signature is not considered, it may be confused with other waveform distortions, especially in the modeling of overlapping signals in next generation ground-based GW detectors. The observation of one of these curves and its associated parameters may help (1) constrain the orbital distance $d_{\rm LS}$ of sources, especially around low-mass MBHs at the center of star clusters and galaxies, (2) additionally estimate the mass $M_{{\rm MBH},z}$ of these MBHs, and (3) infer the orbital inclination of the binary. Simultaneously obtaining $d_{\rm LS}$ and $M_{{\rm MBH},z}$ through self-lensing can help constrain the astrophysical environments where GW signals come from.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 18:38:47 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:13:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ubach",
"Helena",
""
]
] |
2512.08966
|
Anton Alexa
|
Anton Alexa
|
A Dynamical Approach to the Berezin--Li--Yau Inequality
|
Corrected typos; improved the proof of rigidity in Lemma 3.2 and Proposition 5.3; minor consistency fixes
| null | null | null |
math.DG math.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We develop a dynamical method for proving the sharp Berezin--Li--Yau inequality. The approach is based on the volume-preserving mean curvature flow and a new monotonicity principle for the Riesz mean $R_\Lambda(\Omega_t)$. For convex domains we show that $R_\Lambda$ is monotone non-decreasing along the flow. The key input is a geometric correlation inequality between the boundary spectral density $Q_\Lambda$ and the mean curvature $H$, established in all dimensions: in $d=2$ via circular symmetrization, and in $d\ge 3$ via the boundary Weyl expansion together with the Laugesen--Morpurgo trace minimization principle. Since the flow converges smoothly to the ball, the monotonicity implies the sharp Berezin--Li--Yau bound for every smooth convex domain. As an application, we obtain a sharp dynamical Ces\`aro--P\'olya inequality for eigenvalue averages.
|
[
{
"version": "v1",
"created": "Mon, 01 Dec 2025 11:35:01 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 01:13:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Alexa",
"Anton",
""
]
] |
2512.09015
|
Luke Merrick
|
DatologyAI: Luke Merrick, Alex Fang, Aldo Carranza, Alvin Deng, Amro Abbas, Brett Larsen, Cody Blakeney, Darren Teh, David Schwab, Fan Pan, Haakon Mongstad, Haoli Yin, Jack Urbanek, Jason Lee, Jason Telanoff, Josh Wills, Kaleigh Mentzer, Paul Burstein, Parth Doshi, Paul Burnstein, Pratyush Maini, Ricardo Monti, Rishabh Adiga, Scott Loftin, Siddharth Joshi, Spandan Das, Tony Jiang, Vineeth Dorna, Zhengping Wang, Bogdan Gaza, Ari Morcos, Matthew Leavitt
|
Luxical: High-Speed Lexical-Dense Text Embeddings
|
9 pages, 6 figures (v2 fixes typos only)
| null | null | null |
cs.CL cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Frontier language model quality increasingly hinges on our ability to organize web-scale text corpora for training. Today's dominant tools trade off speed and flexibility: lexical classifiers (e.g., FastText) are fast but limited to producing classification output scores, while the vector-valued outputs of transformer text embedding models flexibly support numerous workflows (e.g., clustering, classification, and retrieval) but are computationally expensive to produce. We introduce Luxical, a library for high-speed "lexical-dense" text embeddings that aims to recover the best properties of both approaches for web-scale text organization. Luxical combines sparse TF--IDF features, a small ReLU network, and a knowledge distillation training regimen to approximate large transformer embedding models at a fraction of their operational cost. In this technical report, we describe the Luxical architecture and training objective and evaluate a concrete Luxical model in two disparate applications: a targeted webcrawl document retrieval test and an end-to-end language model data curation task grounded in text classification. In these tasks we demonstrate speedups ranging from 3x to 100x over varying-sized neural baselines, and comparable to FastText model inference during the data curation task. On these evaluations, the tested Luxical model illustrates favorable compute/quality trade-offs for large-scale text organization, matching the quality of neural baselines. Luxical is available as open-source software at https://github.com/datologyai/luxical.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 18:58:44 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 17:14:51 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"DatologyAI",
"",
""
],
[
":",
"",
""
],
[
"Merrick",
"Luke",
""
],
[
"Fang",
"Alex",
""
],
[
"Carranza",
"Aldo",
""
],
[
"Deng",
"Alvin",
""
],
[
"Abbas",
"Amro",
""
],
[
"Larsen",
"Brett",
""
],
[
"Blakeney",
"Cody",
""
],
[
"Teh",
"Darren",
""
],
[
"Schwab",
"David",
""
],
[
"Pan",
"Fan",
""
],
[
"Mongstad",
"Haakon",
""
],
[
"Yin",
"Haoli",
""
],
[
"Urbanek",
"Jack",
""
],
[
"Lee",
"Jason",
""
],
[
"Telanoff",
"Jason",
""
],
[
"Wills",
"Josh",
""
],
[
"Mentzer",
"Kaleigh",
""
],
[
"Burstein",
"Paul",
""
],
[
"Doshi",
"Parth",
""
],
[
"Burnstein",
"Paul",
""
],
[
"Maini",
"Pratyush",
""
],
[
"Monti",
"Ricardo",
""
],
[
"Adiga",
"Rishabh",
""
],
[
"Loftin",
"Scott",
""
],
[
"Joshi",
"Siddharth",
""
],
[
"Das",
"Spandan",
""
],
[
"Jiang",
"Tony",
""
],
[
"Dorna",
"Vineeth",
""
],
[
"Wang",
"Zhengping",
""
],
[
"Gaza",
"Bogdan",
""
],
[
"Morcos",
"Ari",
""
],
[
"Leavitt",
"Matthew",
""
]
] |
2512.09086
|
Zeyu Deng
|
Xinyu Qi, Zeyu Deng, Shaun Alexander Macdonald, Liying Li, Chen Wang, Muhammad Ali Imran, Philip G. Zhao
|
Inferring Operator Emotions from a Motion-Controlled Robotic Arm
| null |
IEEE IoT 2025
| null | null |
cs.RO cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A remote robot operator's affective state can significantly impact the resulting robot's motions leading to unexpected consequences, even when the user follows protocol and performs permitted tasks. The recognition of a user operator's affective states in remote robot control scenarios is, however, underexplored. Current emotion recognition methods rely on reading the user's vital signs or body language, but the devices and user participation these measures require would add limitations to remote robot control. We demonstrate that the functional movements of a remote-controlled robotic avatar, which was not designed for emotional expression, can be used to infer the emotional state of the human operator via a machine-learning system. Specifically, our system achieved 83.3$\%$ accuracy in recognizing the user's emotional state expressed by robot movements, as a result of their hand motions. We discuss the implications of this system on prominent current and future remote robot operation and affective robotic contexts.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 19:57:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Qi",
"Xinyu",
""
],
[
"Deng",
"Zeyu",
""
],
[
"Macdonald",
"Shaun Alexander",
""
],
[
"Li",
"Liying",
""
],
[
"Wang",
"Chen",
""
],
[
"Imran",
"Muhammad Ali",
""
],
[
"Zhao",
"Philip G.",
""
]
] |
2512.09100
|
Karl Svozil
|
Karl Svozil
|
Certified Private Quantum Time Ticks Away Faster than Any Classical Clock
|
6 pages, 2 figures, added a section on certified private quantum time
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce the concept of an entangled clock, where the flow of time is operationally defined by the discrete registration of measurement outcomes on a singlet state. Comparing the synchronization rate of two such clocks against classical models, we find that at obtuse relative angles, the quantum clock ticks strictly faster than this classical benchmark. We further propose a protocol for Certified Private Time, adapting device-independent randomness certification to the temporal domain; this guarantees a sovereign timeline that ticks away faster than any local realistic mechanism allows.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 20:33:04 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:19:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Svozil",
"Karl",
""
]
] |
2512.09111
|
Yuji Takubo
|
Yuji Takubo, Arpit Dwivedi, Sukeerth Ramkumar, Luis A. Pabon, Daniele Gammelli, Marco Pavone, Simone D'Amico
|
Semantic Trajectory Generation for Goal-Oriented Spacecraft Rendezvous
|
28 pages, 12 figures. Submitted to AIAA SCITECH 2026
| null | null | null |
cs.RO cs.AI math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reliable real-time trajectory generation is essential for future autonomous spacecraft. While recent progress in nonconvex guidance and control is paving the way for onboard autonomous trajectory optimization, these methods still rely on extensive expert input (e.g., waypoints, constraints, mission timelines, etc.), which limits the operational scalability in real rendezvous missions. This paper introduces SAGES (Semantic Autonomous Guidance Engine for Space), a trajectory-generation framework that translates natural-language commands into spacecraft trajectories that reflect high-level intent while respecting nonconvex constraints. Experiments in two settings -- fault-tolerant proximity operations with continuous-time constraint enforcement and a free-flying robotic platform -- demonstrate that SAGES reliably produces trajectories aligned with human commands, achieving over 90% semantic-behavioral consistency across diverse behavior modes. Ultimately, this work marks an initial step toward language-conditioned, constraint-aware spacecraft trajectory generation, enabling operators to interactively guide both safety and behavior through intuitive natural-language commands with reduced expert burden.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 20:53:16 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:52:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Takubo",
"Yuji",
""
],
[
"Dwivedi",
"Arpit",
""
],
[
"Ramkumar",
"Sukeerth",
""
],
[
"Pabon",
"Luis A.",
""
],
[
"Gammelli",
"Daniele",
""
],
[
"Pavone",
"Marco",
""
],
[
"D'Amico",
"Simone",
""
]
] |
2512.09160
|
Craig Roberts
|
Peng Cheng, Langtian Liu, Ya Lu and Craig D. Roberts
|
Insights into Nucleon Resonances via Continuum Schwinger Function Methods
|
24 pages, 10 figures. To appear in a special issue of Acta Physica Polonica B, which celebrates the 90th birthday of L. David Roper
| null | null |
NJU-INP 108-25
|
hep-ph hep-ex hep-lat nucl-ex nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The first baryon resonance was discovered in the early 1950s. The Roper resonance joined the collection ten years later. Today, many baryon resonances are known and more are being discovered. As baryons, these states are the most fundamental three-body systems in Nature. They must all be understood, not just the isolated ground state nucleon. This contribution sketches applications of continuum Schwinger function methods to the baryon resonance problem. Whilst spectroscopy is of value, particular emphasis is placed on resonance electroproduction because transition form factors extracted from electroproduction data provide a keen tool for revealing resonance structure.
|
[
{
"version": "v1",
"created": "Tue, 09 Dec 2025 22:19:15 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:47:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cheng",
"Peng",
""
],
[
"Liu",
"Langtian",
""
],
[
"Lu",
"Ya",
""
],
[
"Roberts",
"Craig D.",
""
]
] |
2512.09237
|
Xiaoya Wang
|
Xiaoya Wang, Richard J. Cook, Yeying Zhu, Tugba Akkaya-Hocagil, R. Colin Carter, Sandra W. Jacobson, Joseph L. Jacobson, Louise M. Ryan
|
Prenatal alcohol exposure and child cognition: semi-continuous exposures, causal inference and evidence synthesis
|
arXiv admin note: This version has been removed by arXiv administrators as the submitter did not have the right to agree to the license at the time of submission
| null | null | null |
stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
We address the challenge of causal inference status and the dose-response effects with a semi-continuous exposure. A two-stage approach is proposed using estimating equation for multiple outcomes with large sample properties derived for the resulting estimators. Homogeneity tests are developed to assess whether causal effects of exposure status and the dose-response effects are the same across multiple outcomes. A global homogeneity test is also developed to assess whether the effect of exposure status (exposed/not exposed) and the dose-response effect of the continuous exposure level are each equal across all outcomes. The methods of estimation and testing are rigorously evaluated in simulation studies and applied to a motivating study on the effects of prenatal alcohol exposure on childhood cognition defined by executive function (EF), academic achievement in math, and learning and memory (LM).
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 01:54:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Xiaoya",
""
],
[
"Cook",
"Richard J.",
""
],
[
"Zhu",
"Yeying",
""
],
[
"Akkaya-Hocagil",
"Tugba",
""
],
[
"Carter",
"R. Colin",
""
],
[
"Jacobson",
"Sandra W.",
""
],
[
"Jacobson",
"Joseph L.",
""
],
[
"Ryan",
"Louise M.",
""
]
] |
2512.09316
|
Marios Papamichalis Dr
|
Marios Papamichalis, Nicholas Christakis and Feng Fu
|
Group Cooperation Diverges onto Durable Low versus High Paths: Public Goods Experiments in 134 Honduran Villages
|
This is the initial version of the manuscript. The presentation of figures, tables, and analyses may be revised in future versions to better align with the requirements and scope of the target journal
| null | null | null |
stat.AP stat.OT
|
http://creativecommons.org/licenses/by/4.0/
|
We performed large, lab-in-the-field experiment (2,591 participants across 134 Honduran villages; ten rounds) and tracked how contribution behavior unfolds in fixed, anonymous groups of size five. Contribution separates early into two durable paths, one low and one high, with rare convergence thereafter. High-path players can be identified with strong accuracy early on. Groups that begin with an early majority of above-norm contributors (about 60%) are very likely finish high. The empirical finding of a bifurcation, consistent with the theory, shows that early, high contributions by socially central people steer groups onto, and help keep them on, a high-cooperation path.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 05:03:20 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:11:54 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Papamichalis",
"Marios",
""
],
[
"Christakis",
"Nicholas",
""
],
[
"Fu",
"Feng",
""
]
] |
2512.09317
|
Galen Wilkerson
|
Galen J. Wilkerson
|
Functional Percolation: A Perspective on Criticality of Form and Function
|
6 pages, 6 figures
| null | null | null |
physics.soc-ph cond-mat.stat-mech cs.AI physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding the physical constraints and minimal conditions that enable information processing in extended systems remains a central challenge across disciplines, from neuroscience and artificial intelligence to social and physical networks. Here we study how network connectivity both limits and enables information processing by analyzing random networks across the structural percolation transition. Using cascade-mediated dynamics as a minimal and universal mechanism for propagating state-dependent responses, we examine structural, functional, and information-theoretic observables as functions of mean degree in Erdos-Renyi networks. We find that the emergence of a giant connected component coincides with a sharp transition in realizable information processing: complex input-output response functions become accessible, functional diversity increases rapidly, output entropy rises, and directed information flow quantified by transfer entropy extends beyond local neighborhoods. These coincident transitions define a regime of functional percolation, referring to a sharp expansion of the space of realizable input-output functions at the structural percolation transition. Near criticality, networks exhibit a Pareto-optimal tradeoff between functional complexity and diversity, suggesting that percolation criticality provides a universal organizing principle for information processing in systems with local interactions and propagating influences.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 05:05:10 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:55:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wilkerson",
"Galen J.",
""
]
] |
2512.09335
|
Seonghwa Choi
|
Seonghwa Choi, Moonkyeong Choi, Mingyu Jang, Jaekyung Kim, Jianfei Cai, Wen-Huang Cheng, and Sanghoon Lee
|
Relightable and Dynamic Gaussian Avatar Reconstruction from Monocular Video
|
8 pages, 9 figures, published in ACM MM 2025
|
In Proceedings of the 33rd ACM International Conference on Multimedia. 2025. p. 7405-7414
|
10.1145/3746027.3754851
| null |
cs.CV cs.MM
|
http://creativecommons.org/licenses/by/4.0/
|
Modeling relightable and animatable human avatars from monocular video is a long-standing and challenging task. Recently, Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) methods have been employed to reconstruct the avatars. However, they often produce unsatisfactory photo-realistic results because of insufficient geometrical details related to body motion, such as clothing wrinkles. In this paper, we propose a 3DGS-based human avatar modeling framework, termed as Relightable and Dynamic Gaussian Avatar (RnD-Avatar), that presents accurate pose-variant deformation for high-fidelity geometrical details. To achieve this, we introduce dynamic skinning weights that define the human avatar's articulation based on pose while also learning additional deformations induced by body motion. We also introduce a novel regularization to capture fine geometric details under sparse visual cues. Furthermore, we present a new multi-view dataset with varied lighting conditions to evaluate relight. Our framework enables realistic rendering of novel poses and views while supporting photo-realistic lighting effects under arbitrary lighting conditions. Our method achieves state-of-the-art performance in novel view synthesis, novel pose rendering, and relighting.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 05:51:59 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:18:41 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Choi",
"Seonghwa",
""
],
[
"Choi",
"Moonkyeong",
""
],
[
"Jang",
"Mingyu",
""
],
[
"Kim",
"Jaekyung",
""
],
[
"Cai",
"Jianfei",
""
],
[
"Cheng",
"Wen-Huang",
""
],
[
"Lee",
"Sanghoon",
""
]
] |
2512.09343
|
Ashik E Rasul
|
Ashik E Rasul, Humaira Tasnim, Ji Yu Kim, Young Hyun Lim, Scott Schmitz, Bruce W. Jo, Hyung-Jin Yoon
|
Development and Testing for Perception Based Autonomous Landing of a Long-Range QuadPlane
| null | null | null | null |
cs.RO cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
QuadPlanes combine the range efficiency of fixed-wing aircraft with the maneuverability of multi-rotor platforms for long-range autonomous missions. In GPS-denied or cluttered urban environments, perception-based landing is vital for reliable operation. Unlike structured landing zones, real-world sites are unstructured and highly variable, requiring strong generalization capabilities from the perception system. Deep neural networks (DNNs) provide a scalable solution for learning landing site features across diverse visual and environmental conditions. While perception-driven landing has been shown in simulation, real-world deployment introduces significant challenges. Payload and volume constraints limit high-performance edge AI devices like the NVIDIA Jetson Orin Nano, which are crucial for real-time detection and control. Accurate pose estimation during descent is necessary, especially in the absence of GPS, and relies on dependable visual-inertial odometry. Achieving this with limited edge AI resources requires careful optimization of the entire deployment framework. The flight characteristics of large QuadPlanes further complicate the problem. These aircraft exhibit high inertia, reduced thrust vectoring, and slow response times further complicate stable landing maneuvers. This work presents a lightweight QuadPlane system for efficient vision-based autonomous landing and visual-inertial odometry, specifically developed for long-range QuadPlane operations such as aerial monitoring. It describes the hardware platform, sensor configuration, and embedded computing architecture designed to meet demanding real-time, physical constraints. This establishes a foundation for deploying autonomous landing in dynamic, unstructured, GPS-denied environments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 06:02:25 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:44:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Rasul",
"Ashik E",
""
],
[
"Tasnim",
"Humaira",
""
],
[
"Kim",
"Ji Yu",
""
],
[
"Lim",
"Young Hyun",
""
],
[
"Schmitz",
"Scott",
""
],
[
"Jo",
"Bruce W.",
""
],
[
"Yoon",
"Hyung-Jin",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.