Search is not available for this dataset
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 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s] | authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2411.10227 | Pablo Rosillo-Rodes | Pablo Rosillo-Rodes, Maxi San Miguel and David Sanchez | Entropy and type-token ratio in gigaword corpora | 15 pages, 10 figures, 8 tables | null | null | null | cs.CL cs.IR physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | There are different ways of measuring diversity in complex systems. In
particular, in language, lexical diversity is characterized in terms of the
type-token ratio and the word entropy. We here investigate both diversity
metrics in six massive linguistic datasets in English, Spanish, and Turkish,
consisting of books, news articles, and tweets. These gigaword corpora
correspond to languages with distinct morphological features and differ in
registers and genres, thus constituting a varied testbed for a quantitative
approach to lexical diversity. We unveil an empirical functional relation
between entropy and type-token ratio of texts of a given corpus and language,
which is a consequence of the statistical laws observed in natural language.
Further, in the limit of large text lengths we find an analytical expression
for this relation relying on both Zipf and Heaps laws that agrees with our
empirical findings.
| [
{
"version": "v1",
"created": "Fri, 15 Nov 2024 14:40:59 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 22:21:27 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Rosillo-Rodes",
"Pablo",
""
],
[
"Miguel",
"Maxi San",
""
],
[
"Sanchez",
"David",
""
]
] |
2411.10781 | Shitong Shao | Shitong Shao and Zikai Zhou and Tian Ye and Lichen Bai and Zhiqiang Xu
and Zeke Xie | Bag of Design Choices for Inference of High-Resolution Masked Generative
Transformer | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Text-to-image diffusion models (DMs) develop at an unprecedented pace,
supported by thorough theoretical exploration and empirical analysis.
Unfortunately, the discrepancy between DMs and autoregressive models (ARMs)
complicates the path toward achieving the goal of unified vision and language
generation. Recently, the masked generative Transformer (MGT) serves as a
promising intermediary between DM and ARM by predicting randomly masked image
tokens (i.e., masked image modeling), combining the efficiency of DM with the
discrete token nature of ARM. However, we find that the comprehensive analyses
regarding the inference for MGT are virtually non-existent, and thus we aim to
present positive design choices to fill this gap. We propose and redesign a set
of enhanced inference techniques tailored for MGT, providing a detailed
analysis of their performance. Additionally, we explore several DM-based
approaches aimed at accelerating the sampling process on MGT. Extensive
experiments and empirical analyses on the recent SOTA MGT, such as MaskGIT and
Meissonic lead to concrete and effective design choices, and these design
choices can be merged to achieve further performance gains. For instance, in
terms of enhanced inference, we achieve winning rates of approximately 70%
compared to vanilla sampling on HPS v2 with Meissonic-1024x1024.
| [
{
"version": "v1",
"created": "Sat, 16 Nov 2024 11:51:33 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 13:51:10 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Shao",
"Shitong",
""
],
[
"Zhou",
"Zikai",
""
],
[
"Ye",
"Tian",
""
],
[
"Bai",
"Lichen",
""
],
[
"Xu",
"Zhiqiang",
""
],
[
"Xie",
"Zeke",
""
]
] |
2411.11689 | Nikolaos Fytas G. | Dimitrios Mataragkas, Alexandros Vasilopoulos, Nikolaos G. Fytas, and
Dong-Hee Kim | Tricriticality and finite-size scaling in the triangular Blume-Capel
ferromagnet | 16 pages, 15 figures, 1 table, version accepted for publication in
Physical Review Research | Physical Review Research 7, 013214 (2025) | 10.1103/PhysRevResearch.7.013214 | null | cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We report on numerical simulations of the two-dimensional spin-$1$
Blume-Capel ferromagnet embedded in a triangular lattice. Utilizing a range of
Monte Carlo and finite-size scaling techniques, we explore several critical
aspects along the crystal field--temperature ($\Delta, T$) transition line.
Wang-Landau simulations measuring the joint density of states in combination
with the method of field mixing allow us to probe the phase coexistence curve
in high resolution, determining the tricritical point $(\Delta_{\rm t}, T_{\rm
t})$ with improved accuracy and verifying the tricritical exponents. Extensive
multicanonical simulations identifying transition points across the phase
diagram characterize the Ising universality class for $\Delta < \Delta_{\rm t}$
with precise determination of thermal and magnetic critical exponents expected
in the second-order regime. On the other hand, for $\Delta > \Delta_{\rm t}$, a
finite-size scaling analysis is dedicated to revealing the first-order
signature in the surface tension that linearly increases upon lowering the
temperature deeper into the first-order transition regime. Finally, a
comprehensive picture of the phase diagram for the model is presented,
collecting transition points obtained from the combined numerical approach in
this study and previous estimates in the literature.
| [
{
"version": "v1",
"created": "Mon, 18 Nov 2024 16:12:21 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Jan 2025 07:24:31 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Mataragkas",
"Dimitrios",
""
],
[
"Vasilopoulos",
"Alexandros",
""
],
[
"Fytas",
"Nikolaos G.",
""
],
[
"Kim",
"Dong-Hee",
""
]
] |
2411.13261 | Shu-Min Zhao | Song Gao, Shu-Min Zhao, Long Ruan, Xing-Xing Dong and Tai-Fu Feng | A 95 GeV Higgs Boson in the $U(1)_X$ model | 23 pages, 5 figures | null | null | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The CMS and ATLAS Collaborations have recently reported their findings based
on the comprehensive run 2 dataset, detailing their searches for a light Higgs
boson with a mass of approximately 95 GeV. We investigate the excesses observed
in the $\gamma\gamma$, $\tau{\bar\tau}$ and $b{\bar b}$ data at approximately
95 GeV in the $U(1)_X$ model. Additionally, it also mixes with the SM-like
Higgs boson. Research indicates that, in this model, identifying this singlet
Higgs state as the lightest Higgs boson holds tremendous potential for
explaining the excess observed at approximately 95 GeV. In our calculations, we
maintain the masses of the lightest and next-to-lightest Higgs bosons at
approximately 95 GeV and 125 GeV, respectively. The study finds that the
theoretical predictions for the signal strengths $\mu(h_{95})_{\gamma\gamma}$
$\mu(h_{95})_{\tau{\bar\tau}}$ and $\mu(h_{95})_{b{\bar b}}$ in the $U(1)_X$
model align well with the excesses observed by CMS.
| [
{
"version": "v1",
"created": "Wed, 20 Nov 2024 12:26:20 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 01:38:44 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Gao",
"Song",
""
],
[
"Zhao",
"Shu-Min",
""
],
[
"Ruan",
"Long",
""
],
[
"Dong",
"Xing-Xing",
""
],
[
"Feng",
"Tai-Fu",
""
]
] |
2411.13641 | Benoit Laurent | Wen-Yuan Ai, Benoit Laurent, Jorinde van de Vis | Bounds on the bubble wall velocity | 47 pages, 8 figures; v2 matches published version | JHEP 02 (2025) 119 | 10.1007/JHEP02(2025)119 | CERN-TH-2024-198, KCL-PH-TH/2024-57 | hep-ph astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | Determining the bubble wall velocity in first-order phase transitions is a
challenging task, requiring the solution of (coupled) equations of motion for
the scalar field and Boltzmann equations for the particles in the plasma. The
collision terms appearing in the Boltzmann equation present a prominent source
of uncertainty as they are often known only at leading log accuracy. In this
paper, we derive upper and lower bounds on the wall velocity, corresponding to
the local thermal equilibrium and ballistic limits. These bounds are completely
independent of the collision terms.
For the ballistic approximation, we argue that the inhomogeneous plasma
temperature and velocity distributions across the bubble wall should be taken
into account. This way, the hydrodynamic obstruction previously observed in
local thermal equilibrium is also present for the ballistic approximation. This
is essential for the ballistic approximation to provide a lower bound on the
wall velocity. We use a model-independent approach to study the behaviour of
the limiting wall velocities as a function of a few generic parameters, and we
test our developments in the singlet extended Standard Model.
| [
{
"version": "v1",
"created": "Wed, 20 Nov 2024 19:00:00 GMT"
},
{
"version": "v2",
"created": "Mon, 24 Feb 2025 16:06:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Ai",
"Wen-Yuan",
""
],
[
"Laurent",
"Benoit",
""
],
[
"van de Vis",
"Jorinde",
""
]
] |
2411.13984 | Ryota Fujii | Ryota Fujii and Makoto Sakaguchi | A Comment on the Higher-Spin Gauge Models | 6 pages, published version | Mod. Phys. Lett. A (2025) 2550007 | 10.1142/S0217732325500075 | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We examine the higher-spin gauge models which are free on-shell and whose
cubic vertex is BRST-exact. We show that these are free off-shell as well as
on-shell. The key equation for this relates the derivative of the total
deformed action $S(g)$ with respect to the deformation parameter $g$ to an
$S(g)$-exact term.
| [
{
"version": "v1",
"created": "Thu, 21 Nov 2024 09:47:40 GMT"
},
{
"version": "v2",
"created": "Thu, 16 Jan 2025 00:48:13 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Fujii",
"Ryota",
""
],
[
"Sakaguchi",
"Makoto",
""
]
] |
2411.14117 | Egor Nuzhin | Egor E. Nuzhin and Nikolai V. Brilliantov | Umbrella Reinforcement Learning -- computationally efficient tool for
hard non-linear problems | null | null | 10.1016/j.cnsns.2024.108583 | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | We report a novel, computationally efficient approach for solving hard
nonlinear problems of reinforcement learning (RL). Here we combine umbrella
sampling, from computational physics/chemistry, with optimal control methods.
The approach is realized on the basis of neural networks, with the use of
policy gradient. It outperforms, by computational efficiency and implementation
universality, all available state-of-the-art algorithms, in application to hard
RL problems with sparse reward, state traps and lack of terminal states. The
proposed approach uses an ensemble of simultaneously acting agents, with a
modified reward which includes the ensemble entropy, yielding an optimal
exploration-exploitation balance.
| [
{
"version": "v1",
"created": "Thu, 21 Nov 2024 13:34:36 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Nuzhin",
"Egor E.",
""
],
[
"Brilliantov",
"Nikolai V.",
""
]
] |
2411.14302 | Seong-Ho Shinn | Seong-Ho Shinn, Adolfo del Campo | Electrodynamics of Vortices in Quasi-2D Scalar Bose-Einstein Condensates | 17 pages, 1 figure | Phys. Rev. Research 7, 013217 (2025) | 10.1103/PhysRevResearch.7.013217 | null | cond-mat.quant-gas cond-mat.stat-mech math-ph math.MP physics.plasm-ph quant-ph | http://creativecommons.org/licenses/by/4.0/ | In two spatial dimensions, vortex-vortex interactions approximately vary with
the logarithm of the inter-vortex distance, making it possible to describe an
ensemble of vortices as a Coulomb gas. We introduce a duality between vortices
in a quasi-two-dimensional (quasi-2D) scalar Bose-Einstein condensates (BEC)
and effective Maxwell's electrodynamics. Specifically, we address the general
scenario of inhomogeneous, time-dependent BEC number density with dissipation
or rotation. Starting from the Gross-Pitaevskii equation (GPE), which describes
the mean-field dynamics of a quasi-2D scalar BEC without dissipation, we show
how to map vortices in a quasi-2D scalar BEC to 2D electrodynamics beyond the
point-vortex approximation, even when dissipation is present or in a rotating
system. The physical meaning of this duality is discussed.
| [
{
"version": "v1",
"created": "Thu, 21 Nov 2024 16:55:50 GMT"
},
{
"version": "v2",
"created": "Sun, 24 Nov 2024 10:37:16 GMT"
},
{
"version": "v3",
"created": "Mon, 2 Dec 2024 17:05:26 GMT"
},
{
"version": "v4",
"created": "Wed, 26 Feb 2025 14:05:53 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Shinn",
"Seong-Ho",
""
],
[
"del Campo",
"Adolfo",
""
]
] |
2411.14562 | Sara Torelli | Andreas Leopold Knutsen, Sara Torelli | Brill-Noether loci of pencils with prescribed ramification on moduli of
curves and on Severi varieties on $K3$ surfaces | 23 pages - minor changes | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Under the assumption that the adjusted Brill-Noether number
$\widetilde{\rho}$ is at least $-g$, we prove that the Brill-Noether loci in
$\mathcal{M}_{g,n}$ of pointed curves carrying pencils with prescribed
ramification at the marked points have a component of the expected codimension
with pointed curves having Brill-Noether varieties of pencils of the minimal
dimension. As an application, the map from the Hurwitz scheme to
$\mathcal{M}_g$ is dominant if $n+\widetilde{\rho} \geq 0$ and generically
finite otherwise, settling a variation of a classical problem of Zariski.
In the second part of the paper, we study the analogous loci of curves in
Severi varieties on $K3$ surfaces, proving existence of curves with non-general
behaviour from the point of view of Brill-Noether theory. This extends previous
results of Ciliberto and the first named author to the ramified case. We apply
these results to study correspondences and cycles on $K3$ surfaces in relation
to Beauville-Voisin points and constant cycle curves.
| [
{
"version": "v1",
"created": "Thu, 21 Nov 2024 20:24:38 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:29:26 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Knutsen",
"Andreas Leopold",
""
],
[
"Torelli",
"Sara",
""
]
] |
2411.14998 | Roberto Emparan | Roberto Emparan, Mikel Sanchez-Garitaonandia, Marija Toma\v{s}evi\'c | String Theory in a Pinch: Resolving the Gregory-Laflamme Singularity | 34 pages, 9 figures. v2: revised and improved discussion of
stability. 38 pages, 10 figures. v3: matches published version | JHEP 02 (2025) 104 | 10.1007/JHEP02(2025)104 | CPHT-RR082.112024 | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Thin enough black strings are unstable to growing ripples along their length,
eventually pinching and forming a naked singularity on the horizon. We
investigate how string theory can resolve this singularity. First, we study the
string-scale version of the static non-uniform black strings that branch off at
the instability threshold: "string-ball strings", which are linearly extended,
self-gravitating configurations of string balls obtained in the
Horowitz-Polchinski (HP) approach to near-Hagedorn string states. We construct
non-uniform HP strings in spatial dimensions $d\leq 6$ and show that, as the
inhomogeneity increases, they approach localized HP balls. We also examine the
thermodynamic properties of the different phases in the canonical and
microcanonical ensembles. We find that, for a sufficiently small mass, the
uniform HP string will be stable and not evolve into a non-uniform or localized
configuration. Building on these results and independent evidence from the
evolution of the black string instability with $\alpha'$ corrections, we
propose that, at least in $d=4,5$, string theory slows and eventually halts the
pinching evolution at a classically stable stringy neck. In $d\geq 6$ this
transition is likely to occur into a puffed-up string ball. The system then
enters a slower phase in which the neck gradually evaporates into radiation. We
discuss this scenario as a framework for understanding how string theory
resolves the formation of naked singularities.
| [
{
"version": "v1",
"created": "Fri, 22 Nov 2024 15:13:45 GMT"
},
{
"version": "v2",
"created": "Fri, 13 Dec 2024 18:46:34 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 17:51:56 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Emparan",
"Roberto",
""
],
[
"Sanchez-Garitaonandia",
"Mikel",
""
],
[
"Tomašević",
"Marija",
""
]
] |
2411.16032 | Adam Rycerz | Adam Rycerz | Quantum transport of Dirac fermions in selected graphene nanosystems
away from the charge-neutrality point | Typos corrected; Refs added | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Peculiar electronic properties of graphene, including the universal dc
conductivity and the pseudodiffusive shot noise, are usually attributed to a
small vicinity of the charge-neutrality point, away from which electron's
effective mass raises, and nanostructures in graphene start to behave similarly
to familiar Sharvin contacts in semiconducting heterostructures hosting
two-dimensional electron gas. Using the effective Dirac equation for low-energy
excitations it can be shown that, as long as abrupt potential steps separate
the sample area from the leads, graphene-specific features can be identified
even relatively far from the charge-neutrality point. Namely, the conductance
is reduced, comparing to the standard Sharvin value, whereas the shot noise is
amplified. Here, we confront the results of earlier analytic considerations
with numerical simulations of quantum transport on the honeycomb lattice, for
selected systems for which considerations starting from the Dirac equation
cannot be directly adapted. For a wedge-shape constriction with zigzag edges
and approximately square shape of the narrowest section, the transport
characteristics can be tuned from graphene-specific sub-Sharvin values to
standard Sharvin values, depending on whether the electrostatic potential
profile in the narrowest section is rectangular or smooth. The half-Corbino
disk with rectangular potential profile exhibits both the conductance and the
noise close to the sub-Sharvin values. For a circular quantum dot with two
narrow openings and irregular edges, the conductance is close to the Sharvin
value, and the Fano factor approaches the value of $F=1/4$. This suggests that,
in experimental attempt to verify the predictions for sub-Sharvin transport
regime, one should focus rather on nanosystems, for which the scatterings on
edges are insignificant next to the scatterings on sample-lead interfaces.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 01:21:20 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Dec 2024 19:25:38 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Feb 2025 18:24:57 GMT"
},
{
"version": "v4",
"created": "Thu, 27 Feb 2025 18:45:35 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Rycerz",
"Adam",
""
]
] |
2411.16084 | Shuyan Cheng | Shuyan Cheng, Yishu Wei, Yiliang Zhou, Zihan Xu, Drew N Wright, Jinze
Liu, Yifan Peng | Deciphering genomic codes using advanced NLP techniques: a scoping
review | null | null | 10.1093/jamia/ocaf029 | null | q-bio.GN cs.AI | http://creativecommons.org/licenses/by/4.0/ | Objectives: The vast and complex nature of human genomic sequencing data
presents challenges for effective analysis. This review aims to investigate the
application of Natural Language Processing (NLP) techniques, particularly Large
Language Models (LLMs) and transformer architectures, in deciphering genomic
codes, focusing on tokenization, transformer models, and regulatory annotation
prediction. The goal of this review is to assess data and model accessibility
in the most recent literature, gaining a better understanding of the existing
capabilities and constraints of these tools in processing genomic sequencing
data.
Methods: Following Preferred Reporting Items for Systematic Reviews and
Meta-Analyses (PRISMA) guidelines, our scoping review was conducted across
PubMed, Medline, Scopus, Web of Science, Embase, and ACM Digital Library.
Studies were included if they focused on NLP methodologies applied to genomic
sequencing data analysis, without restrictions on publication date or article
type.
Results: A total of 26 studies published between 2021 and April 2024 were
selected for review. The review highlights that tokenization and transformer
models enhance the processing and understanding of genomic data, with
applications in predicting regulatory annotations like transcription-factor
binding sites and chromatin accessibility.
Discussion: The application of NLP and LLMs to genomic sequencing data
interpretation is a promising field that can help streamline the processing of
large-scale genomic data while also providing a better understanding of its
complex structures. It has the potential to drive advancements in personalized
medicine by offering more efficient and scalable solutions for genomic
analysis. Further research is also needed to discuss and overcome current
limitations, enhancing model transparency and applicability.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 04:35:56 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Cheng",
"Shuyan",
""
],
[
"Wei",
"Yishu",
""
],
[
"Zhou",
"Yiliang",
""
],
[
"Xu",
"Zihan",
""
],
[
"Wright",
"Drew N",
""
],
[
"Liu",
"Jinze",
""
],
[
"Peng",
"Yifan",
""
]
] |
2411.16231 | Ilia Pavlov | I. Pavlov, A. Chaikovskaia, D. Karlovets | Angular momentum effects in neutron decay | 16 pages, 17 figures | Phys. Rev. C 111, 024619 (2025) | 10.1103/PhysRevC.111.024619 | null | hep-ph quant-ph | http://creativecommons.org/licenses/by/4.0/ | We investigate the intriguing phenomenon of beta decay of a free neutron in a
non-plane-wave(structured) state. Our analysis covers three types of states:
unpolarized vortex (Bessel) neutrons that possess nonzero orbital angular
momentum (OAM), Laguerre-Gaussian wave packets, and spin-correlated OAM
(spin-orbit) states characterized by unique polarization patterns. These states
are of particular interest as they have recently been generated in neutron
optics experiments and have promising applications in studies of quantum
magnetic materials. The spectral-angular distributions (SAD) of the emitted
electrons and protons are examined. We show that the high sensitivity of the
protons SAD to the structure of the neutron wave packet can be used as a tool
to extract the distinctive features of the non-plane-wave neutron states.
Furthermore, we demonstrate that the angular distribution of the emitted
particles serves as a reflection of the spatial symmetries inherent to the
neutron wave packet.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 09:44:28 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:21:33 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Pavlov",
"I.",
""
],
[
"Chaikovskaia",
"A.",
""
],
[
"Karlovets",
"D.",
""
]
] |
2411.16349 | Roman Belousov | Irem Topal, Alexander Cherevko, Yuri Bugay, Maxim Shishlenin, Jean
Barbier, Deniz Eroglu, \'Edgar Rold\'an, Roman Belousov | Machine learning for cerebral blood vessels' malformations | 15 pages, 6 main figures, 5 supplementary figures, 2 supplementary
tables | null | null | null | cs.LG cond-mat.stat-mech q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | Cerebral aneurysms and arteriovenous malformations are life-threatening
hemodynamic pathologies of the brain. While surgical intervention is often
essential to prevent fatal outcomes, it carries significant risks both during
the procedure and in the postoperative period, making the management of these
conditions highly challenging. Parameters of cerebral blood flow, routinely
monitored during medical interventions or with modern noninvasive
high-resolution imaging methods, could potentially be utilized in machine
learning-assisted protocols for risk assessment and therapeutic prognosis. To
this end, we developed a linear oscillatory model of blood velocity and
pressure for clinical data acquired from neurosurgical operations. Using the
method of Sparse Identification of Nonlinear Dynamics (SINDy), the parameters
of our model can be reconstructed online within milliseconds from a short time
series of the hemodynamic variables. The identified parameter values enable
automated classification of the blood-flow pathologies by means of logistic
regression, achieving an accuracy of 73 \%}. Our results demonstrate the
potential of this model for both diagnostic and prognostic applications,
providing a robust and interpretable framework for assessing cerebral blood
vessel conditions.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 12:58:00 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 15:57:27 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Topal",
"Irem",
""
],
[
"Cherevko",
"Alexander",
""
],
[
"Bugay",
"Yuri",
""
],
[
"Shishlenin",
"Maxim",
""
],
[
"Barbier",
"Jean",
""
],
[
"Eroglu",
"Deniz",
""
],
[
"Roldán",
"Édgar",
""
],
[
"Belousov",
"Roman",
""
]
] |
2411.16537 | Chan Hee Song | Chan Hee Song, Valts Blukis, Jonathan Tremblay, Stephen Tyree, Yu Su,
Stan Birchfield | RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language
Models for Robotics | null | CVPR 2025 | null | null | cs.CV cs.AI cs.CL cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatial understanding is a crucial capability for robots to make grounded
decisions based on their environment. This foundational skill enables robots
not only to perceive their surroundings but also to reason about and interact
meaningfully within the world. In modern robotics, these capabilities are taken
on by visual language models, and they face significant challenges when applied
to spatial reasoning context due to their training data sources. These sources
utilize general-purpose image datasets, and they often lack sophisticated
spatial scene understanding capabilities. For example, the datasets do not
address reference frame comprehension - spatial relationships require clear
contextual understanding, whether from an ego-centric, object-centric, or
world-centric perspective, which allow for effective real-world interaction. To
address this issue, we introduce RoboSpatial, a large-scale spatial
understanding dataset consisting of real indoor and tabletop scenes captured as
3D scans and egocentric images, annotated with rich spatial information
relevant to robotics. The dataset includes 1M images, 5K 3D scans, and 3M
annotated spatial relationships, with paired 2D egocentric images and 3D scans
to make it both 2D and 3D ready. Our experiments show that models trained with
RoboSpatial outperform baselines on downstream tasks such as spatial affordance
prediction, spatial relationship prediction, and robotics manipulation.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 16:21:34 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Song",
"Chan Hee",
""
],
[
"Blukis",
"Valts",
""
],
[
"Tremblay",
"Jonathan",
""
],
[
"Tyree",
"Stephen",
""
],
[
"Su",
"Yu",
""
],
[
"Birchfield",
"Stan",
""
]
] |
2411.16547 | Cesar Augusto Ipanaque Zapata Prof. | Cesar A. Ipanaque Zapata, Josu\'e A. Aguirre Enciso and Wilman
Francisco Cuba Ramos | (Injective) hom-complexity between graphs | 32 pages. It is a completely new version which is pitched more
clearly at the readership and direction of some specific journal. We also
discuss a connection between the (injective) hom-complexity and several
well-known covering numbers. Comments are welcome | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the notion of hom-complexity, $\text{C}(G;H)$, for two graphs $G$
and $H$, along with basic results for this numerical invariant. This invariant
$\text{C}(G;H)$ is a number that measures the \aspas{complexity} of the
question: when is there a homomorphism $G\to H$? More precisely,
$\text{C}(G;H)$ is the least number $\ell$ such that there are $\ell$ different
subgraphs $G_j$ of $G$ such that $G=G_1\cup\cdots\cup G_\ell$, and for each
$G_j$, there is a homomorphism $G_j\to H$. Likewise, we introduce the notion of
injective hom-complexity, $\text{IC}(G;H)$. The (injective) hom-complexity is a
graph invariant. Additionally, these invariants provide an obstruction to the
existence of homomorphisms. We explore the sub-additivity of (injective)
hom-complexity and study products.
We establish a relation between $\text{C}(G;H)$ and the chromatic numbers
$\chi(G)$ and $\chi(H)$. Specifically, we present lower and upper bounds for
the hom-complexity in terms of the chromatic number and the clique number. We
provide the formula $\text{C}(G;H)=\lceil\log_{\chi(H)}\chi(G)\rceil$ whenever
$\omega(H)=\chi(H)$. For example, we obtain
$\text{C}(G;K_n)=\lceil\log_{n}\chi(G)\rceil$. We also demonstrate a triangular
inequality for (injective) hom-complexity. Moreover, we discuss a connection
between the (injective) hom-complexity and several well-known covering numbers.
For instance, we provide a lower bound for the clique covering number in terms
of the injective hom-complexity. Additionally, we show that the hom-complexity
$\mathrm{C}(G;K_{\ell})$ coincides with the $\ell$-particity of $G$, and the
hom-complexity $\mathrm{C}(K_n;K_{2})$ coincides with the bipartite dimension
of $K_n$. As a consequence, we recover the well-known formulas
$\beta_\ell(G)=\lceil\log_{\ell}\chi(G)\rceil$ and
$\mathrm{d}(K_n)=\lceil\log_{2}n\rceil$.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 16:31:17 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 02:55:08 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zapata",
"Cesar A. Ipanaque",
""
],
[
"Enciso",
"Josué A. Aguirre",
""
],
[
"Ramos",
"Wilman Francisco Cuba",
""
]
] |
2411.16917 | James Marshall | James A. R. Marshall and Andrew B. Barron | Are Transformers Truly Foundational for Robotics? | null | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Generative Pre-Trained Transformers (GPTs) are hyped to revolutionize
robotics. Here we question their utility. GPTs for autonomous robotics demand
enormous and costly compute, excessive training times and (often) offboard
wireless control. We contrast GPT state of the art with how tiny insect brains
have achieved robust autonomy with none of these constraints. We highlight
lessons that can be learned from biology to enhance the utility of GPTs in
robotics.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 20:35:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 10:46:58 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Marshall",
"James A. R.",
""
],
[
"Barron",
"Andrew B.",
""
]
] |
2411.17732 | Naveed Bhatti | Abdur-Rahman Ibrahim Sayyid-Ali, Abdul Rafay, Muhammad Abdullah
Soomro, Muhammad Hamad Alizai, Naveed Anwar Bhatti | CheckMate: LLM-Powered Approximate Intermittent Computing | Accepted in SenSys 2025 | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Batteryless IoT systems face energy constraints exacerbated by checkpointing
overhead. Approximate computing offers solutions but demands manual expertise,
limiting scalability. This paper presents CheckMate, an automated framework
leveraging LLMs for context-aware code approximations. CheckMate integrates
validation of LLM-generated approximations to ensure correct execution and
employs Bayesian optimization to fine-tune approximation parameters
autonomously, eliminating the need for developer input. Tested across six IoT
applications, it reduces power cycles by up to 60% with an accuracy loss of
just 8%, outperforming semi-automated tools like ACCEPT in speedup and
accuracy. CheckMate's results establish it as a robust, user-friendly tool and
a foundational step toward automated approximation frameworks for intermittent
computing.
| [
{
"version": "v1",
"created": "Fri, 22 Nov 2024 11:18:26 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:05:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Sayyid-Ali",
"Abdur-Rahman Ibrahim",
""
],
[
"Rafay",
"Abdul",
""
],
[
"Soomro",
"Muhammad Abdullah",
""
],
[
"Alizai",
"Muhammad Hamad",
""
],
[
"Bhatti",
"Naveed Anwar",
""
]
] |
2411.17771 | Xinyu Zhang | Xinyu Zhang, Lingling Zhang, Yanrui Wu, Muye Huang, Wenjun Wu, Bo Li,
Shaowei Wang, Jun Liu | DiagramQG: A Dataset for Generating Concept-Focused Questions from
Diagrams | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Visual Question Generation (VQG) has gained significant attention due to its
potential in educational applications. However, VQG researches mainly focus on
natural images, neglecting diagrams in educational materials used to assess
students' conceptual understanding. To address this gap, we introduce
DiagramQG, a dataset containing 8,372 diagrams and 19,475 questions across
various subjects. DiagramQG introduces concept and target text constraints,
guiding the model to generate concept-focused questions for educational
purposes. Meanwhile, we present the Hierarchical Knowledge Integration
framework for Diagram Question Generation (HKI-DQG) as a strong baseline. This
framework obtains multi-scale patches of diagrams and acquires knowledge using
a visual language model with frozen parameters. It then integrates knowledge,
text constraints and patches to generate concept-focused questions. We evaluate
the performance of existing VQG models, open-source and closed-source
vision-language models, and HKI-DQG on the DiagramQG dataset. Our HKI-DQG
outperform existing methods, demonstrating that it serves as a strong baseline.
Furthermore, we apply HKI-DQG to four other VQG datasets of natural images,
namely VQG-COCO, K-VQG, OK-VQA and A-OKVQA, achieving state-of-the-art
performance. The dataset and code are available at
https://dxzxy12138.github.io/diagramqg-home.
| [
{
"version": "v1",
"created": "Tue, 26 Nov 2024 08:27:50 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 15:16:17 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zhang",
"Xinyu",
""
],
[
"Zhang",
"Lingling",
""
],
[
"Wu",
"Yanrui",
""
],
[
"Huang",
"Muye",
""
],
[
"Wu",
"Wenjun",
""
],
[
"Li",
"Bo",
""
],
[
"Wang",
"Shaowei",
""
],
[
"Liu",
"Jun",
""
]
] |
2411.18369 | Tianxing Chen | Tianxing Chen, Yao Mu, Zhixuan Liang, Zanxin Chen, Shijia Peng,
Qiangyu Chen, Mingkun Xu, Ruizhen Hu, Hongyuan Zhang, Xuelong Li, Ping Luo | G3Flow: Generative 3D Semantic Flow for Pose-aware and Generalizable
Object Manipulation | Webpage: https://tianxingchen.github.io/G3Flow/, accepted to CVPR
2025 | null | null | null | cs.RO cs.AI cs.CV cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in imitation learning for 3D robotic manipulation have shown
promising results with diffusion-based policies. However, achieving human-level
dexterity requires seamless integration of geometric precision and semantic
understanding. We present G3Flow, a novel framework that constructs real-time
semantic flow, a dynamic, object-centric 3D semantic representation by
leveraging foundation models. Our approach uniquely combines 3D generative
models for digital twin creation, vision foundation models for semantic feature
extraction, and robust pose tracking for continuous semantic flow updates. This
integration enables complete semantic understanding even under occlusions while
eliminating manual annotation requirements. By incorporating semantic flow into
diffusion policies, we demonstrate significant improvements in both
terminal-constrained manipulation and cross-object generalization. Extensive
experiments across five simulation tasks show that G3Flow consistently
outperforms existing approaches, achieving up to 68.3% and 50.1% average
success rates on terminal-constrained manipulation and cross-object
generalization tasks respectively. Our results demonstrate the effectiveness of
G3Flow in enhancing real-time dynamic semantic feature understanding for
robotic manipulation policies.
| [
{
"version": "v1",
"created": "Wed, 27 Nov 2024 14:17:43 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 17:59:12 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Chen",
"Tianxing",
""
],
[
"Mu",
"Yao",
""
],
[
"Liang",
"Zhixuan",
""
],
[
"Chen",
"Zanxin",
""
],
[
"Peng",
"Shijia",
""
],
[
"Chen",
"Qiangyu",
""
],
[
"Xu",
"Mingkun",
""
],
[
"Hu",
"Ruizhen",
""
],
[
"Zhang",
"Hongyuan",
""
],
[
"Li",
"Xuelong",
""
],
[
"Luo",
"Ping",
""
]
] |
2411.18508 | Yiwei Liu | Yiwei Liu, Yihu Yang | Some rigidity results related to the Obata type equation | null | null | null | null | math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $(\Omega^{n+1},g)$ be an $(n + 1)$-dimensional smooth complete connected
Riemannian manifold with compact boundary $\partial\Omega=\Sigma$ and $f$ a
smooth function on $\Omega$ which satisfies the Obata type equation $\nabla^2 f
-fg =0$ with Robin boundary condition $f_{\nu} = cf$, where
$c=\coth{\theta}>1$. In this paper, we provide some rigidity results based on
the warped product structure of $\Omega$ determined by the equation $\nabla^2 f
-fg =0$ and appropriate curvature assumptions. We also apply a similar method
to the Obata type equation $\nabla^2 f +fg =0$ and get a rigidity result on the
standard sphere $\mathbb{S}^{n+1}$.
| [
{
"version": "v1",
"created": "Wed, 27 Nov 2024 16:51:03 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 07:06:27 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Liu",
"Yiwei",
""
],
[
"Yang",
"Yihu",
""
]
] |
2411.19120 | Maxim Dvornikov | Maxim Dvornikov (IZMIRAN) | Quantum field theory treatment of neutrino flavor oscillations in matter | 19 pages in pdflatex, 5 pdf figures; title was changed, text was
revised significantly, version to be published in Phys.Rev.D | null | null | null | hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | We study neutrino oscillations in background matter within the quantum field
theory formalism where neutrino mass eigenstates are virtual particles. In this
case, neutrino mass eigenstates are mixed owing to the interaction with matter.
Assuming that neutrinos are Majorana particles, we find the exact propagators
for massive neutrinos accounting for the interaction with matter by solving the
analog of the Dyson equation. These propagators are used to calculate the
transition probability which coincides with the prediction of the standard
quantum mechanical treatment of neutrino flavor oscillations in uniform matter.
Finally, we analyze the approximations made in our analysis.
| [
{
"version": "v1",
"created": "Thu, 28 Nov 2024 13:10:43 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 12:23:14 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Dvornikov",
"Maxim",
"",
"IZMIRAN"
]
] |
2411.19671 | Xianliang Li | Xianliang Li, Jun Luo, Zhiwei Zheng, Hanxiao Wang, Li Luo, Lingkun
Wen, Linlong Wu, Sheng Xu | On the Performance Analysis of Momentum Method: A Frequency Domain
Perspective | ICLR 2025, 22 pages, 14 figures. Keywords: Momentum Method,
Stochastic Gradient Descent, Z-Transform, Frequency Domain Analysis, Deep
Learning | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Momentum-based optimizers are widely adopted for training neural networks.
However, the optimal selection of momentum coefficients remains elusive. This
uncertainty impedes a clear understanding of the role of momentum in stochastic
gradient methods. In this paper, we present a frequency domain analysis
framework that interprets the momentum method as a time-variant filter for
gradients, where adjustments to momentum coefficients modify the filter
characteristics. Our experiments support this perspective and provide a deeper
understanding of the mechanism involved. Moreover, our analysis reveals the
following significant findings: high-frequency gradient components are
undesired in the late stages of training; preserving the original gradient in
the early stages, and gradually amplifying low-frequency gradient components
during training both enhance performance. Based on these insights, we propose
Frequency Stochastic Gradient Descent with Momentum (FSGDM), a heuristic
optimizer that dynamically adjusts the momentum filtering characteristic with
an empirically effective dynamic magnitude response. Experimental results
demonstrate the superiority of FSGDM over conventional momentum optimizers.
| [
{
"version": "v1",
"created": "Fri, 29 Nov 2024 12:56:43 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Feb 2025 01:34:12 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 08:33:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Li",
"Xianliang",
""
],
[
"Luo",
"Jun",
""
],
[
"Zheng",
"Zhiwei",
""
],
[
"Wang",
"Hanxiao",
""
],
[
"Luo",
"Li",
""
],
[
"Wen",
"Lingkun",
""
],
[
"Wu",
"Linlong",
""
],
[
"Xu",
"Sheng",
""
]
] |
2411.19675 | Edward Skevington | Edward W.G. Skevington and Robert M. Dorrell | Gravity current energetics and particle suspension | null | null | null | null | physics.flu-dyn | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Gravity currents are a ubiquitous density driven flow occurring in both the
natural environment and in industry. They include: seafloor turbidity currents,
primary vectors of sediment, nutrient and pollutant transport; cold fronts; and
hazardous gas spills. However, while the energetics are critical for their
evolution and particle suspension, they are included in system scale models
only crudely, so we cannot yet predict and explain the dynamics and run-out of
such real-world flows. Herein, a novel depth-averaged framework is developed to
capture the evolution of volume, concentration, momentum, and turbulent kinetic
energy from direct integrals of the full governing equations. For the first
time, we show the connection between the vertical profiles, the evolution of
the depth-averaged flow, and the energetics. The viscous dissipation of
mean-flow energy near the bed makes a leading order contribution, and an
energetic approach to entrainment captures detrainment of fluid through
particle settling. These observations allow a reconsideration of particle
suspension, advancing over 50-years of research. We find that the new
formulation can describe the full evolution of a shallow dilute current, with
the accuracy depending primarily on closures for the profiles and source terms.
Critically, this enables accurate and computationally efficient hazard risk
analysis and earth surface modelling.
| [
{
"version": "v1",
"created": "Fri, 29 Nov 2024 13:04:12 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 14:35:49 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Skevington",
"Edward W. G.",
""
],
[
"Dorrell",
"Robert M.",
""
]
] |
2412.00273 | Julian Siegl | Julian Siegl, Anton Bleibaum, Wen Wan, Marcin Kurpas, John Schliemann,
Miguel M. Ugeda, Magdalena Marganska, and Milena Grifoni | Friedel oscillations and chiral superconductivity in monolayer NbSe$_2$ | 12 pages, 4 figures | null | null | null | cond-mat.supr-con cond-mat.mes-hall | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In 1965 Kohn and Luttinger proposed a genuine electronic mechanism for
superconductivity. Despite the bare electrostatic interaction between two
electrons being repulsive, in a metal electron-hole fluctuations can give rise
to Friedel oscillations of the screened Coulomb potential. Cooper pairing among
the electrons then emerges when taking advantage of the attractive regions. The
nature of the leading pairing mechanism in some two-dimensional transition
metal dichalcogenides is still debated. Focusing on NbSe$_2$, we show that
superconductivity can be induced by the Coulomb interaction when accounting for
screening effects on the trigonal lattice with multiple orbitals. Using ab
initio-based tight-binding parametrizations for the relevant low-energy
d-bands, we evaluate the screened interaction microscopically, in a scheme
including Bloch overlaps. In the direct space, we find long-range Friedel
oscillations alternating in sign, a key to the Kohn-Luttinger mechanism. The
momentum-resolved gap equations predict two degenerate solutions at the
critical temperature Tc, signaling the unconventional nature of the pairing.
Their complex linear combination, i.e., a chiral gap with p-like symmetry,
provides the ground state of the system. Our prediction of a fully gapped
chiral phase well below Tc is in excellent agreement with the spectral function
extracted from tunneling spectroscopy measurements of single-layer NbSe$_2$.
| [
{
"version": "v1",
"created": "Fri, 29 Nov 2024 22:16:07 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:22:47 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Siegl",
"Julian",
""
],
[
"Bleibaum",
"Anton",
""
],
[
"Wan",
"Wen",
""
],
[
"Kurpas",
"Marcin",
""
],
[
"Schliemann",
"John",
""
],
[
"Ugeda",
"Miguel M.",
""
],
[
"Marganska",
"Magdalena",
""
],
[
"Grifoni",
"Milena",
""
]
] |
2412.00303 | Reggie Pantig | Reggie C. Pantig, Gaetano Lambiase, Ali \"Ovg\"un, Nikko John Leo S.
Lobos | Spacetime-curvature induced uncertainty principle: linking the
large-structure global effects to the local black hole physics | 13 pages, 0 figures and tables. Published in Physics of the Dark
Universe | Phys. Dark Univ. 47, 2025, 101817 | 10.1016/j.dark.2025.101817 | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper links the advanced formulation of the Generalized Uncertainty
Principle, termed the Asymptotic Generalized Extended Uncertainty Principle
(AGEUP), to the corpuscular framework to derive the AGEUP-inspired black hole
metric. The former incorporates spacetime curvature effects to explore black
hole dynamics under quantum gravitational corrections, while the latter is a
view that black holes are Bose-Einstein condensates of weakly interacting
gravitons. In a particular case, the phenomenological union between the AGEUP
with cosmological constant $\Lambda$ to the corpuscular framework enabled a
black hole metric that has a scaled mass, which depends on $\Lambda$ and the
Planck length $l_{\rm Pl}$. Interesting implications occur, such as the maximum
limit for mass $M$ where $\Lambda$ ceases to influence the black hole. Another
is the derived value of the modulation factor of the EUP term, $\alpha$, if the
large-scale fundamental length is defined solely as the cosmological horizon.
Additional analyses were done through the shadow and deflection angle
phenomena, deriving constraints on the quantum gravity modulation parameter
$\beta$. Constraints from the Event Horizon Telescope (EHT) and Very Long
Baseline Interferometry (VLBI) are discussed as avenues for verifying
AGEUP-related deviations in black hole shadow radius and deflection angles,
offering potential observational evidence of quantum gravitational effects at
astrophysical scales. The findings suggest that AGEUP could be instrumental in
providing hints on the quantum gravity nature of black holes, particularly in
high-energy astrophysical contexts. By linking local black hole physics with
large-scale curvature effects, AGEUP paves the way for further research at the
intersection of quantum gravity and cosmology, with implications for
observational astrophysics and the fundamental structure of spacetime.
| [
{
"version": "v1",
"created": "Sat, 30 Nov 2024 01:14:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 05:03:29 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Pantig",
"Reggie C.",
""
],
[
"Lambiase",
"Gaetano",
""
],
[
"Övgün",
"Ali",
""
],
[
"Lobos",
"Nikko John Leo S.",
""
]
] |
2412.00588 | Mahmudul Islam | Saifuddin Zafar, Mashaekh Tausif Ehsan, Sourav Das Suvro, Mohammad
Nasim Hasan, and Mahmudul Islam | Enhancing creep resistance in refractory high-entropy alloys: role of
grain size and local chemical order | 23 pages, 6 figures | null | null | null | cond-mat.mtrl-sci physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Refractory high-entropy alloys (RHEAs) are a promising class of materials
with potential applications in extreme environments, where the dominant failure
mode is thermal creep. The design of these alloys, therefore, requires an
understanding of how their microstructure and local chemical distribution
affect creep behavior. In this study, we performed high-fidelity atomistic
simulations using machine-learning interatomic potentials to explore the creep
deformation of MoNbTaW RHEAs under a wide range of stress and temperature
conditions. We parametrized grain size and local chemical order (LCO) to
investigate the effects of these two important design variables, which are
controllable during the alloy fabrication process. Our investigation revealed
that resistance to creep deformation is enhanced by larger grain sizes and
higher levels of LCO. This study highlights the importance of utilizing LCO in
conjunction with other microstructural properties when designing RHEAs for
extreme environment applications.
| [
{
"version": "v1",
"created": "Sat, 30 Nov 2024 21:07:47 GMT"
},
{
"version": "v2",
"created": "Sun, 8 Dec 2024 18:26:30 GMT"
},
{
"version": "v3",
"created": "Tue, 7 Jan 2025 02:59:17 GMT"
},
{
"version": "v4",
"created": "Thu, 27 Feb 2025 16:23:44 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zafar",
"Saifuddin",
""
],
[
"Ehsan",
"Mashaekh Tausif",
""
],
[
"Suvro",
"Sourav Das",
""
],
[
"Hasan",
"Mohammad Nasim",
""
],
[
"Islam",
"Mahmudul",
""
]
] |
2412.00792 | Maolin Bo | Yaorui Tan and Maolin Bo | Electronic Structure, mass fluctuation, and Localized Bond Properties of
two-dimensional double-layer transition metal chalcogenide MX$_2$ (M = Mo, W;
X = S, Se, Te) Calculated Based on Density Functional Theory and BBC model | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study systematically investigates the electronic structure and bonding
properties of two-dimensional bilayer transition metal chalcogenides MX2 (M =
Mo, W; X = S, Se, Te) using density functional theory calculations. By
analyzing band gaps, deformation bond energies, and non-Hermitian bonding
characteristics across various MX2 compounds, we comprehensively examine their
electronic properties and chemical bonding behavior. The results reveal that
charge transfer plays a crucial role in electron mass fluctuations, with
topological geometric analysis further confirming the impact of mass variations
on atomic bonding and electronic states. These findings provide a theoretical
foundation for advancing the application of these materials.
| [
{
"version": "v1",
"created": "Sun, 1 Dec 2024 12:38:14 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Jan 2025 08:08:37 GMT"
},
{
"version": "v3",
"created": "Sun, 12 Jan 2025 11:38:22 GMT"
},
{
"version": "v4",
"created": "Sat, 25 Jan 2025 06:19:38 GMT"
},
{
"version": "v5",
"created": "Thu, 27 Feb 2025 13:41:16 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Tan",
"Yaorui",
""
],
[
"Bo",
"Maolin",
""
]
] |
2412.00904 | Hans Rabus | Hans Rabus, Leo Thomas | On a revised concept of an event that allows linking nanodosimetry and
microdosimetry in nanometric sites with macroscopic dosimetry | 39 pages, 8 Figs. plus 11 Suppl. Figs | Radiation Physics and Chemistry 232; 112640 (2025). Online
23.02.2025 | 10.1016/j.radphyschem.2025.112640 | null | physics.med-ph | http://creativecommons.org/licenses/by/4.0/ | This work reviews the concepts of an event used in micro- and nanodosimetry
and analyzes how single event distributions could theoretically be derived from
probability distributions related to interactions of the primary particle which
produce secondary electrons. It is shown that the corresponding mathematical
expressions of conditional ionization cluster size distributions are alike
those for the single event frequency distribution of energy imparted,
particularly when all tracks are considered which intersect the volume in which
interactions of the primary particle can result in energy deposits in the site.
Track structure simulations of proton with energies between 1 MeV and 100 MeV
are used to study how the occurrence of events depends on site size, beam
radius, and proton energy. The range of impact parameters of particle tracks
that contribute to energy imparted in a site appears not to depend on whether
any energy deposits or only energy deposits by ionizations are considered.
Since there is no longer a one-to-one correspondence between tracks passing a
site and the occurrence of an event, it is proposed to use the fluence for
which on average one event occurs as a substitute for single events. For
protons, the product of this fluence and the site cross section or the average
number of tracks necessary for an event shows an interesting dependence on site
size and particle energy with asymptotic values close to unity for large sites
and proton energies below 10 MeV. For a proton energy of 1 MeV, a minimum of
the number of tracks is observed for sites between 5 nm and 10 nm diameter. The
relative differences between the numbers of track per event on average obtained
with different options of Geant4-DNA are in the order of 10 % and illustrate
the need for further investigations into cross-section datasets and their
uncertainties.
| [
{
"version": "v1",
"created": "Sun, 1 Dec 2024 17:37:56 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Feb 2025 15:42:11 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Rabus",
"Hans",
""
],
[
"Thomas",
"Leo",
""
]
] |
2412.01150 | Steven Dillmann | Steven Dillmann, Rafael Mart\'inez-Galarza, Roberto Soria, Rosanne Di
Stefano, Vinay L. Kashyap | Representation Learning for Time-Domain High-Energy Astrophysics:
Discovery of Extragalactic Fast X-ray Transient XRT 200515 | 25 pages, submitted to Monthly Notices of the Royal Astronomical
Society, presented at the 2023 Conference on Machine Learning in Astronomical
Surveys (ML-IAP/CCA-2023) | Monthly Notices of the Royal Astronomical Society, Volume 537,
Issue 2, February 2025 | 10.1093/mnras/stae2808 | null | astro-ph.HE astro-ph.IM cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel representation learning method for downstream tasks such
as anomaly detection and unsupervised transient classification in high-energy
datasets. This approach enabled the discovery of a new fast X-ray transient
(FXT) in the Chandra archive, XRT 200515, a needle-in-the-haystack event and
the first Chandra FXT of its kind. Recent serendipitous breakthroughs in X-ray
astronomy, including FXTs from binary neutron star mergers and an extragalactic
planetary transit candidate, highlight the need for systematic transient
searches in X-ray archives. We introduce new event file representations, E-t
Maps and E-t-dt Cubes, designed to capture both temporal and spectral
information, effectively addressing the challenges posed by variable-length
event file time series in machine learning applications. Our pipeline extracts
low-dimensional, informative features from these representations using
principal component analysis or sparse autoencoders, followed by clustering in
the embedding space with DBSCAN. New transients are identified within
transient-dominant clusters or through nearest-neighbor searches around known
transients, producing a catalog of 3,539 candidates (3,427 flares and 112
dips). XRT 200515 exhibits unique temporal and spectral variability, including
an intense, hard <10 s initial burst followed by spectral softening in an ~800
s oscillating tail. We interpret XRT 200515 as either the first giant magnetar
flare observed at low X-ray energies or the first extragalactic Type I X-ray
burst from a faint LMXB in the LMC. Our method extends to datasets from other
observatories such as XMM-Newton, Swift-XRT, eROSITA, Einstein Probe, and
upcoming missions like AXIS.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 05:48:31 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Dillmann",
"Steven",
""
],
[
"Martínez-Galarza",
"Rafael",
""
],
[
"Soria",
"Roberto",
""
],
[
"Di Stefano",
"Rosanne",
""
],
[
"Kashyap",
"Vinay L.",
""
]
] |
2412.01702 | Roberto Capecelatro | Roberto Capecelatro, Marco Marciani, Gabriele Campagnano, Procolo
Lucignano | Andreev non-Hermitian Hamiltonian for open Josephson junctions from
Green's functions | 23 pages, 17 figures, review | Phys. Rev. B 111, 06451 - Published 27 February, 2025 Vol. 111,
Iss. 6 - 1 February 2025 | 10.1103/PhysRevB.111.064517 | null | cond-mat.supr-con cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the transport properties of open Josephson junctions (JJs)
through a minimal effective non-Hermitian (NH) approach derived from the
equilibrium Green's function (GF) formalism. Specifically, we consider a JJ
with a quantum dot barrier coupled to a normal metal reservoir. The coupling
introduces an imaginary self-energy term in the JJ Hamiltonian which can be
naturally accounted for in the NH formalism. While most approaches to similar
problems work with the full junction Hamiltonian we propose a scheme for
deriving an effective NH Hamiltonian for the Andreev levels only, that we
compute from the singular part of the barrier GF. To establish the range of
applicability of this NH model we benchmark our results for both the dot
density of states and the supercurrent against exact GF predictions in
different transport regimes. We find that, as a rule of thumb, the Andreev NH
description is accurate when the spectral overlap between the Andreev bound
states (ABS) and the near-gap continuum states is negligible, i.e. when the ABS
energies lie sufficiently far from the superconducting gap relative to their
line-width. This method not only highlights the effective physics of the JJ but
also offers a scalable framework for studying large-size devices.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 16:49:26 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 17:56:06 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Capecelatro",
"Roberto",
""
],
[
"Marciani",
"Marco",
""
],
[
"Campagnano",
"Gabriele",
""
],
[
"Lucignano",
"Procolo",
""
]
] |
2412.01928 | Sumeet Motwani | Sumeet Ramesh Motwani, Chandler Smith, Rocktim Jyoti Das, Rafael
Rafailov, Ivan Laptev, Philip H. S. Torr, Fabio Pizzati, Ronald Clark,
Christian Schroeder de Witt | MALT: Improving Reasoning with Multi-Agent LLM Training | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) often produce answers with a single
chain-of-thought, which restricts their ability to explore reasoning paths or
self-correct flawed outputs in complex tasks. In this paper, we introduce MALT
(Multi-Agent LLM Training), a novel post-training strategy that divides the
reasoning process into generation, verification, and refinement steps using a
sequential pipeline of heterogeneous agents. During data generation, each agent
is repeatedly sampled to form a multi-agent search tree, where final outputs
are graded against ground-truth data. We then apply value iteration to
propagate reward signals back to each role-conditioned model, automatically
producing multi-agent post-training data without human or teacher-model
supervision. Our off-policy approach allows each agent to specialize by
learning from correct and incorrect trajectories, ultimately improving the
end-to-end reasoning chain. On MATH, GSM8K, and CSQA, MALT surpasses the same
baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40%
respectively, making it an important advance towards multi-agent cooperative
training.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 19:30:36 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:25:02 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Motwani",
"Sumeet Ramesh",
""
],
[
"Smith",
"Chandler",
""
],
[
"Das",
"Rocktim Jyoti",
""
],
[
"Rafailov",
"Rafael",
""
],
[
"Laptev",
"Ivan",
""
],
[
"Torr",
"Philip H. S.",
""
],
[
"Pizzati",
"Fabio",
""
],
[
"Clark",
"Ronald",
""
],
[
"de Witt",
"Christian Schroeder",
""
]
] |
2412.02877 | Prathami Divakar Kamath | Alin Marin Elena (1), Prathami Divakar Kamath (2,3), Th\'eo Jaffrelot
Inizan (2,3), Andrew S. Rosen (2,3), Federica Zanca (1), Kristin A. Persson
(2,3) ((1) Scientific Computing Department, Science and Technology Facilities
Council, Cheshire, UK, (2) Department of Materials Science and Engineering,
University of California Berkley, California, USA, (3) Materials Sciences
Division, Lawrence Berkeley National Laboratory, California, USA) | Machine Learned Potential for High-Throughput Phonon Calculations of
Metal-Organic Frameworks | 17 pages, 8 Figures | null | null | null | cond-mat.mtrl-sci physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | Metal-organic frameworks (MOFs) are highly porous and versatile materials
studied extensively for applications such as carbon capture and water
harvesting. However, computing phonon-mediated properties in MOFs, like thermal
expansion and mechanical stability, remains challenging due to the large number
of atoms per unit cell, making traditional Density Functional Theory (DFT)
methods impractical for high-throughput screening. Recent advances in machine
learning potentials have led to foundation atomistic models, such as MACE-MP-0,
that accurately predict equilibrium structures but struggle with phonon
properties of MOFs. In this work, we developed a workflow for computing phonons
in MOFs within the quasi-harmonic approximation with a fine-tuned MACE model,
MACE-MP-MOF0. The model was trained on a curated dataset of 127 representative
and diverse MOFs. The fine-tuned MACE-MP-MOF0 improves the accuracy of phonon
density of states and corrects the imaginary phonon modes of MACE-MP-0,
enabling high-throughput phonon calculations with state-of-the-art precision.
The model successfully predicts thermal expansion and bulk moduli in agreement
with DFT and experimental data for several well-known MOFs. These results
highlight the potential of MACE-MP-MOF0 in guiding MOF design for applications
in energy storage and thermoelectrics.
| [
{
"version": "v1",
"created": "Tue, 3 Dec 2024 22:25:10 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Dec 2024 21:05:59 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 17:40:20 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Elena",
"Alin Marin",
""
],
[
"Kamath",
"Prathami Divakar",
""
],
[
"Inizan",
"Théo Jaffrelot",
""
],
[
"Rosen",
"Andrew S.",
""
],
[
"Zanca",
"Federica",
""
],
[
"Persson",
"Kristin A.",
""
]
] |
2412.03139 | Itai Epstein | Tomer Eini, M. F. C. Martins Quintela, J. C. G. Henriques, R. M.
Ribeiro, Yarden Mazor, N. M. R. Peres and Itai Epstein | Electrically Tunable Interband Collective Excitations in Biased Bilayer
and Trilayer Graphene | null | null | null | null | cond-mat.mes-hall physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Collective excitations of charged particles under the influence of an
electromagnetic field give rise to a rich variety of hybrid light-matter
quasiparticles with unique properties. In metals, intraband collective response
manifested by negative permittivity leads to plasmon-polaritons with extreme
field confinement, wavelength squeezing, and potentially low propagation
losses. In contrast, photons in semiconductors commonly couple to interband
collective response in the form of exciton polaritons, which give rise to
completely different polaritonic properties, described by a superposition of
the photon and exciton and an anti-crossing of the eigenstates. In this work,
we identify the existence of plasmon-like collective excitations originating
from the interband excitonic response of biased bilayer and trilayer graphene,
in the form of graphene-exciton-polaritons (GEPs). We find that GEPs possess
electrically tunable polaritonic properties and discover that such excitations
follow a universal dispersion law for all surface polaritons in 2D excitonic
systems. Accounting for nonlocal corrections to the excitonic response, we find
that the GEPs exhibit confinement factors that can exceed those of graphene
plasmons, and with moderate losses. These predictions of plasmon-like interband
collective excitations in biased graphene systems open up new research avenues
for tunable polaritonic phenomena based on excitonic systems, and the ability
to control and manipulate such phenomena at the atomic scale.
| [
{
"version": "v1",
"created": "Wed, 4 Dec 2024 09:04:53 GMT"
},
{
"version": "v2",
"created": "Tue, 17 Dec 2024 16:05:44 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 07:20:48 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Eini",
"Tomer",
""
],
[
"Quintela",
"M. F. C. Martins",
""
],
[
"Henriques",
"J. C. G.",
""
],
[
"Ribeiro",
"R. M.",
""
],
[
"Mazor",
"Yarden",
""
],
[
"Peres",
"N. M. R.",
""
],
[
"Epstein",
"Itai",
""
]
] |
2412.03844 | Jiaqi Gu | Jingyu Lin, Jiaqi Gu, Lubin Fan, Bojian Wu, Yujing Lou, Renjie Chen,
Ligang Liu, Jieping Ye | HybridGS: Decoupling Transients and Statics with 2D and 3D Gaussian
Splatting | Accpeted by CVPR 2025. Project page:
https://gujiaqivadin.github.io/hybridgs/ Code:
https://github.com/Yeyuqqwx/HybridGS Data:
https://huggingface.co/Eto63277/HybridGS/tree/main | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Generating high-quality novel view renderings of 3D Gaussian Splatting (3DGS)
in scenes featuring transient objects is challenging. We propose a novel hybrid
representation, termed as HybridGS, using 2D Gaussians for transient objects
per image and maintaining traditional 3D Gaussians for the whole static scenes.
Note that, the 3DGS itself is better suited for modeling static scenes that
assume multi-view consistency, but the transient objects appear occasionally
and do not adhere to the assumption, thus we model them as planar objects from
a single view, represented with 2D Gaussians. Our novel representation
decomposes the scene from the perspective of fundamental viewpoint consistency,
making it more reasonable. Additionally, we present a novel multi-view
regulated supervision method for 3DGS that leverages information from
co-visible regions, further enhancing the distinctions between the transients
and statics. Then, we propose a straightforward yet effective multi-stage
training strategy to ensure robust training and high-quality view synthesis
across various settings. Experiments on benchmark datasets show our
state-of-the-art performance of novel view synthesis in both indoor and outdoor
scenes, even in the presence of distracting elements.
| [
{
"version": "v1",
"created": "Thu, 5 Dec 2024 03:20:35 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Dec 2024 04:59:24 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 02:48:54 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Lin",
"Jingyu",
""
],
[
"Gu",
"Jiaqi",
""
],
[
"Fan",
"Lubin",
""
],
[
"Wu",
"Bojian",
""
],
[
"Lou",
"Yujing",
""
],
[
"Chen",
"Renjie",
""
],
[
"Liu",
"Ligang",
""
],
[
"Ye",
"Jieping",
""
]
] |
2412.04141 | Hongshen Xu | Hongshen Xu, Zichen Zhu, Lei Pan, Zihan Wang, Su Zhu, Da Ma, Ruisheng
Cao, Lu Chen, Kai Yu | Reducing Tool Hallucination via Reliability Alignment | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have expanded their capabilities beyond language
generation to interact with external tools, enabling automation and real-world
applications. However, tool hallucinations, where models either select
inappropriate tools or misuse them, pose significant challenges, leading to
erroneous task execution, increased computational costs, and reduced system
reliability. To systematically address this issue, we define and categorize
tool hallucinations into two main types, tool selection hallucination and tool
usage hallucination. To evaluate and mitigate these issues, we introduce
RelyToolBench, which integrates specialized test cases and novel metrics to
assess hallucination-aware task success and efficiency. Finally, we propose
Relign, a reliability alignment framework that expands the tool-use action
space to include indecisive actions, allowing LLMs to defer tool use, seek
clarification, or adjust tool selection dynamically. Through extensive
experiments, we demonstrate that Relign significantly reduces tool
hallucinations, improves task reliability, and enhances the efficiency of LLM
tool interactions.
| [
{
"version": "v1",
"created": "Thu, 5 Dec 2024 13:10:54 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 04:43:54 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Xu",
"Hongshen",
""
],
[
"Zhu",
"Zichen",
""
],
[
"Pan",
"Lei",
""
],
[
"Wang",
"Zihan",
""
],
[
"Zhu",
"Su",
""
],
[
"Ma",
"Da",
""
],
[
"Cao",
"Ruisheng",
""
],
[
"Chen",
"Lu",
""
],
[
"Yu",
"Kai",
""
]
] |
2412.04149 | Xiangyuan Wang | Haitian Zhang, Xiangyuan Wang, Chang Xu, Xinya Wang, Fang Xu, Huai Yu,
Lei Yu, Wen Yang | Frequency-Adaptive Low-Latency Object Detection Using Events and Frames | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fusing Events and RGB images for object detection leverages the robustness of
Event cameras in adverse environments and the rich semantic information
provided by RGB cameras. However, two critical mismatches: low-latency Events
\textit{vs.}~high-latency RGB frames; temporally sparse labels in training
\textit{vs.}~continuous flow in inference, significantly hinder the
high-frequency fusion-based object detection. To address these challenges, we
propose the \textbf{F}requency-\textbf{A}daptive Low-Latency \textbf{O}bject
\textbf{D}etector (FAOD). FAOD aligns low-frequency RGB frames with
high-frequency Events through an Align Module, which reinforces cross-modal
style and spatial proximity to address the Event-RGB Mismatch. We further
propose a training strategy, Time Shift, which enforces the module to align the
prediction from temporally shifted Event-RGB pairs and their original
representation, that is, consistent with Event-aligned annotations. This
strategy enables the network to use high-frequency Event data as the primary
reference while treating low-frequency RGB images as supplementary information,
retaining the low-latency nature of the Event stream toward high-frequency
detection. Furthermore, we observe that these corrected Event-RGB pairs
demonstrate better generalization from low training frequency to higher
inference frequencies compared to using Event data alone. Extensive experiments
on the PKU-DAVIS-SOD and DSEC-Detection datasets demonstrate that our FAOD
achieves SOTA performance. Specifically, in the PKU-DAVIS-SOD Dataset, FAOD
achieves 9.8 points improvement in terms of the mAP in fully paired Event-RGB
data with only a quarter of the parameters compared to SODFormer, and even
maintains robust performance (only a 3 points drop in mAP) under 80$\times$
Event-RGB frequency mismatch.
| [
{
"version": "v1",
"created": "Thu, 5 Dec 2024 13:23:06 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 12:43:29 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zhang",
"Haitian",
""
],
[
"Wang",
"Xiangyuan",
""
],
[
"Xu",
"Chang",
""
],
[
"Wang",
"Xinya",
""
],
[
"Xu",
"Fang",
""
],
[
"Yu",
"Huai",
""
],
[
"Yu",
"Lei",
""
],
[
"Yang",
"Wen",
""
]
] |
2412.04332 | Junfeng Wu | Junfeng Wu, Yi Jiang, Chuofan Ma, Yuliang Liu, Hengshuang Zhao, Zehuan
Yuan, Song Bai, Xiang Bai | Liquid: Language Models are Scalable and Unified Multi-modal Generators | Technical report. Project page:
https://foundationvision.github.io/Liquid/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present Liquid, an auto-regressive generation paradigm that seamlessly
integrates visual comprehension and generation by tokenizing images into
discrete codes and learning these code embeddings alongside text tokens within
a shared feature space for both vision and language. Unlike previous multimodal
large language model (MLLM), Liquid achieves this integration using a single
large language model (LLM), eliminating the need for external pretrained visual
embeddings such as CLIP. For the first time, Liquid uncovers a scaling law that
performance drop unavoidably brought by the unified training of visual and
language tasks diminishes as the model size increases. Furthermore, the unified
token space enables visual generation and comprehension tasks to mutually
enhance each other, effectively removing the typical interference seen in
earlier models. We show that existing LLMs can serve as strong foundations for
Liquid, saving 100x in training costs while outperforming Chameleon in
multimodal capabilities and maintaining language performance comparable to
mainstream LLMs like LLAMA2. Liquid also outperforms models like SD v2.1 and
SD-XL (FID of 5.47 on MJHQ-30K), excelling in both vision-language and
text-only tasks. This work demonstrates that LLMs such as Qwen2.5 and GEMMA2
are powerful multimodal generators, offering a scalable solution for enhancing
both vision-language understanding and generation. The code and models will be
released at https://github.com/FoundationVision/Liquid.
| [
{
"version": "v1",
"created": "Thu, 5 Dec 2024 16:48:16 GMT"
},
{
"version": "v2",
"created": "Thu, 12 Dec 2024 18:08:56 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 16:08:51 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Wu",
"Junfeng",
""
],
[
"Jiang",
"Yi",
""
],
[
"Ma",
"Chuofan",
""
],
[
"Liu",
"Yuliang",
""
],
[
"Zhao",
"Hengshuang",
""
],
[
"Yuan",
"Zehuan",
""
],
[
"Bai",
"Song",
""
],
[
"Bai",
"Xiang",
""
]
] |
2412.04910 | Donald Kougang Yombi | Emmanuel Abbe, Elisabetta Cornacchia, Jan H\k{a}z{\l}a and Donald
Kougang-Yombi | Learning High-Degree Parities: The Crucial Role of the Initialization | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Parities have become a standard benchmark for evaluating learning algorithms.
Recent works show that regular neural networks trained by gradient descent can
efficiently learn degree $k$ parities on uniform inputs for constant $k$, but
fail to do so when $k$ and $d-k$ grow with $d$ (here $d$ is the ambient
dimension). However, the case where $k=d-O_d(1)$ (almost-full parities),
including the degree $d$ parity (the full parity), has remained unsettled. This
paper shows that for gradient descent on regular neural networks, learnability
depends on the initial weight distribution. On one hand, the discrete
Rademacher initialization enables efficient learning of almost-full parities,
while on the other hand, its Gaussian perturbation with large enough constant
standard deviation $\sigma$ prevents it. The positive result for almost-full
parities is shown to hold up to $\sigma=O(d^{-1})$, pointing to questions about
a sharper threshold phenomenon. Unlike statistical query (SQ) learning, where a
singleton function class like the full parity is trivially learnable, our
negative result applies to a fixed function and relies on an initial gradient
alignment measure of potential broader relevance to neural networks learning.
| [
{
"version": "v1",
"created": "Fri, 6 Dec 2024 10:05:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:08:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Abbe",
"Emmanuel",
""
],
[
"Cornacchia",
"Elisabetta",
""
],
[
"Hązła",
"Jan",
""
],
[
"Kougang-Yombi",
"Donald",
""
]
] |
2412.06273 | Dongxu Wei | Dongxu Wei, Zhiqi Li, Peidong Liu | Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View
Scene Reconstruction | Accepted by CVPR2025 | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Prior works employing pixel-based Gaussian representation have demonstrated
efficacy in feed-forward sparse-view reconstruction. However, such
representation necessitates cross-view overlap for accurate depth estimation,
and is challenged by object occlusions and frustum truncations. As a result,
these methods require scene-centric data acquisition to maintain cross-view
overlap and complete scene visibility to circumvent occlusions and truncations,
which limits their applicability to scene-centric reconstruction. In contrast,
in autonomous driving scenarios, a more practical paradigm is ego-centric
reconstruction, which is characterized by minimal cross-view overlap and
frequent occlusions and truncations. The limitations of pixel-based
representation thus hinder the utility of prior works in this task. In light of
this, this paper conducts an in-depth analysis of different representations,
and introduces Omni-Gaussian representation with tailored network design to
complement their strengths and mitigate their drawbacks. Experiments show that
our method significantly surpasses state-of-the-art methods, pixelSplat and
MVSplat, in ego-centric reconstruction, and achieves comparable performance to
prior works in scene-centric reconstruction.
| [
{
"version": "v1",
"created": "Mon, 9 Dec 2024 07:48:15 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 02:53:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Wei",
"Dongxu",
""
],
[
"Li",
"Zhiqi",
""
],
[
"Liu",
"Peidong",
""
]
] |
2412.06958 | Jorge Guevara | Jorge Guevara and Victor Nascimento and Johannes Schmude and Daniel
Salles and Simon Corbeil-L\'etourneau and Madalina Surcel and Dominique
Brunet | Enhancing operational wind downscaling capabilities over Canada:
Application of a Conditional Wasserstein GAN methodology | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Wind downscaling is essential for improving the spatial resolution of weather
forecasts, particularly in operational Numerical Weather Prediction (NWP). This
study advances wind downscaling by extending the DownGAN framework introduced
by Annau et al.,to operational datasets from the Global Deterministic
Prediction System (GDPS) and High-Resolution Deterministic Prediction System
(HRDPS), covering the entire Canadian domain. We enhance the model by
incorporating high-resolution static covariates, such as HRDPS-derived
topography, into a Conditional Wasserstein Generative Adversarial Network with
Gradient Penalty, implemented using a UNET-based generator. Following the
DownGAN framework, our methodology integrates low-resolution GDPS forecasts (15
km, 10-day horizon) and high-resolution HRDPS forecasts (2.5 km, 48-hour
horizon) with Frequency Separation techniques adapted from computer vision.
Through robust training and inference over the Canadian region, we demonstrate
the operational scalability of our approach, achieving significant improvements
in wind downscaling accuracy. Statistical validation highlights reductions in
root mean square error (RMSE) and log spectral distance (LSD) metrics compared
to the original DownGAN. High-resolution conditioning covariates and Frequency
Separation strategies prove instrumental in enhancing model performance. This
work underscores the potential for extending high-resolution wind forecasts
beyond the 48-hour horizon, bridging the gap to the 10-day low resolution
global forecast window.
| [
{
"version": "v1",
"created": "Mon, 9 Dec 2024 20:05:07 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Feb 2025 20:21:13 GMT"
},
{
"version": "v3",
"created": "Wed, 26 Feb 2025 19:34:45 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Guevara",
"Jorge",
""
],
[
"Nascimento",
"Victor",
""
],
[
"Schmude",
"Johannes",
""
],
[
"Salles",
"Daniel",
""
],
[
"Corbeil-Létourneau",
"Simon",
""
],
[
"Surcel",
"Madalina",
""
],
[
"Brunet",
"Dominique",
""
]
] |
2412.07140 | Beilin Chu | Beilin Chu, Xuan Xu, Xin Wang, Yufei Zhang, Weike You, Linna Zhou | FIRE: Robust Detection of Diffusion-Generated Images via
Frequency-Guided Reconstruction Error | 14 pages, 14 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid advancement of diffusion models has significantly improved
high-quality image generation, making generated content increasingly
challenging to distinguish from real images and raising concerns about
potential misuse. In this paper, we observe that diffusion models struggle to
accurately reconstruct mid-band frequency information in real images,
suggesting the limitation could serve as a cue for detecting diffusion model
generated images. Motivated by this observation, we propose a novel method
called Frequency-guided Reconstruction Error (FIRE), which, to the best of our
knowledge, is the first to investigate the influence of frequency decomposition
on reconstruction error. FIRE assesses the variation in reconstruction error
before and after the frequency decomposition, offering a robust method for
identifying diffusion model generated images. Extensive experiments show that
FIRE generalizes effectively to unseen diffusion models and maintains
robustness against diverse perturbations.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 03:02:34 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:33:18 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Chu",
"Beilin",
""
],
[
"Xu",
"Xuan",
""
],
[
"Wang",
"Xin",
""
],
[
"Zhang",
"Yufei",
""
],
[
"You",
"Weike",
""
],
[
"Zhou",
"Linna",
""
]
] |
2412.07234 | Mohammad R. Garousi | Mohammad R. Garousi | Exploring types I and IIA effective actions through T-duality | 8 pages, latex file, no figure; v2: Additional clarifications have
been incorporated into the manuscript | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is well-established that compactification of type I string theory on a
circle $S^{(1)}$ transforms under T-duality into type I' theory, which is the
compactification of type IIA string theory on the orbifold
$\tilde{S}^{(1)}/\mathbb{Z}_2$, where $\mathbb{Z}_2$ combines world-sheet
parity with spacetime parity along the dual circle $\tilde{S}^{(1)}$. We
propose that the classical effective action of type I, when reduced on a
circle, should transform under the Buscher rules into the corresponding
orbifold reduction of the classical effective action of type IIA theory. This
T-duality symmetry can serve as a powerful constraint to determine the bosonic
couplings in the 10-dimensional effective actions of both type I and type IIA
theories, particularly those couplings that remain non-vanishing under orbifold
reduction.
In the classical type IIA effective action at the critical dimension, bosonic
fields appear at the 2-derivative, 8-derivative, and higher-order levels. In
contrast, the type I effective action includes additional 4-derivative,
6-derivative, and higher-order couplings, which arise from the anomalous
Ramond-Ramond (R-R) gauge transformation in the Green-Schwarz mechanism. We
demonstrate that our proposal uniquely fixes the leading 2-derivative couplings
in both type I and type IIA theories, with the exception of the Chern-Simons
coupling in type IIA theory, which vanishes under orbifold reduction.
Additionally, we observe that the higher-derivative couplings in the
10-dimensional type I effective action, stemming from the anomalous R-R gauge
transformation, are exclusively associated with higher-genus contributions.
These terms cannot map to any corresponding terms in the 10-dimensional type
IIA effective action, as recent findings indicate that only classical effective
actions remain invariant under T-duality upon circular reduction.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 06:49:12 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:20:14 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Garousi",
"Mohammad R.",
""
]
] |
2412.07259 | Shaoyu Wang | Shaoyu Wang, Kaiyue Zhao, Dongliang Wei, Przemys{\l}aw Andrzej
Wa{\l}\k{e}ga, Dingmin Wang, Hongming Cai, Pan Hu | Goal-Driven Reasoning in DatalogMTL with Magic Sets | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | DatalogMTL is a powerful rule-based language for temporal reasoning. Due to
its high expressive power and flexible modeling capabilities, it is suitable
for a wide range of applications, including tasks from industrial and financial
sectors. However, due to its high computational complexity, practical reasoning
in DatalogMTL is highly challenging. To address this difficulty, we introduce a
new reasoning method for DatalogMTL which exploits the magic sets technique --
a rewriting approach developed for (non-temporal) Datalog to simulate top-down
evaluation with bottom-up reasoning. We have implemented this approach and
evaluated it on publicly available benchmarks, showing that the proposed
approach significantly and consistently outperformed state-of-the-art reasoning
techniques.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 07:40:37 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Dec 2024 02:24:15 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 14:13:20 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Wang",
"Shaoyu",
""
],
[
"Zhao",
"Kaiyue",
""
],
[
"Wei",
"Dongliang",
""
],
[
"Wałęga",
"Przemysław Andrzej",
""
],
[
"Wang",
"Dingmin",
""
],
[
"Cai",
"Hongming",
""
],
[
"Hu",
"Pan",
""
]
] |
2412.07847 | Steffen Gielen | Steffen Gielen | Hilbert space formalisms for group field theory | 22 pages, 1 figure. Review that makes a few technical points not
easily found in the literature; v2: added some additional comments and
extended some discussions | null | null | null | gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | Group field theory is a background-independent approach to quantum gravity
whose starting point is the definition of a quantum field theory on an
auxiliary group manifold (not interpreted as spacetime, but rather as the
finite-dimensional configuration space of a single "atom" of geometry). Group
field theory models can be seen as an extension of matrix and tensor models by
additional data, and are traditionally defined through a functional integral
whose perturbative expansion generates a sum over discrete geometries. More
recently, some efforts have been directed towards formulations of group field
theory based on a Hilbert space and operators, in particular in applications to
cosmology. This is an attempt to review some of these formulations and their
main ideas, to disentangle these constructions as much as possible from
applications and phenomenology, and to put them into a wider context of quantum
gravity research.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 19:00:08 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:29:42 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Gielen",
"Steffen",
""
]
] |
2412.08058 | Jun Xiao | Sujan Subedi, Wenhao Liu, Wuzhang Fang, Carter Fox, Zixin Zhai, Fan
Fei, Yuan Ping, Bing Lv, Jun Xiao | Colossal terahertz emission with ultrafast tunability based on van der
Waals ferroelectric NbOI$_2$ | null | Advanced Optical Materials (2025) | 10.1002/adom.202403471 | null | cond-mat.mtrl-sci physics.app-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Terahertz (THz) technology is critical for quantum material physics,
biomedical imaging, ultrafast electronics, and next-generation wireless
communications. However, standing in the way of widespread applications is the
scarcity of efficient ultrafast THz sources with on-demand fast modulation and
easy on-chip integration capability. Here we report the discovery of colossal
THz emission from a van der Waals (vdW) ferroelectric semiconductor NbOI$_2$.
Using THz emission spectroscopy, we observe a THz generation efficiency an
order of magnitude higher than that of ZnTe, a standard nonlinear crystal for
ultrafast THz generation. We further uncover the underlying generation
mechanisms associated with its large ferroelectric polarization by studying the
THz emission dependence on excitation wavelength, incident polarization and
fluence. Moreover, we demonstrate the ultrafast coherent amplification and
annihilation of the THz emission and associated coherent phonon oscillations by
employing a double-pump scheme. These findings combined with first-principles
calculations, inform new understanding of the THz light-matter interaction in
emergent vdW ferroelectrics and pave the way to develop high-performance THz
devices on them for quantum materials sensing and ultrafast electronics.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 03:01:55 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Subedi",
"Sujan",
""
],
[
"Liu",
"Wenhao",
""
],
[
"Fang",
"Wuzhang",
""
],
[
"Fox",
"Carter",
""
],
[
"Zhai",
"Zixin",
""
],
[
"Fei",
"Fan",
""
],
[
"Ping",
"Yuan",
""
],
[
"Lv",
"Bing",
""
],
[
"Xiao",
"Jun",
""
]
] |
2412.08521 | Yingxin Li | Yingxin Li, Ye Li, Yuan Meng, Xinzhu Ma, Zihan Geng, Shutao Xia, Zhi
Wang | EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache
Compression Based on Global-Local Importance | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As large language models (LLMs) continue to advance, the demand for higher
quality and faster processing of long contexts across various applications is
growing. KV cache is widely adopted as it stores previously generated key and
value tokens, effectively reducing redundant computations during inference.
However, as memory overhead becomes a significant concern, efficient
compression of KV cache has gained increasing attention. Most existing methods
perform compression from two perspectives: identifying important tokens and
designing compression strategies. However, these approaches often produce
biased distributions of important tokens due to the influence of accumulated
attention scores or positional encoding. Furthermore, they overlook the
sparsity and redundancy across different heads, which leads to difficulties in
preserving the most effective information at the head level. To this end, we
propose EMS to overcome these limitations, while achieving better KV cache
compression under extreme compression ratios. Specifically, we introduce a
Global-Local score that combines accumulated attention scores from both global
and local KV tokens to better identify the token importance. For the
compression strategy, we design an adaptive and unified Evict-then-Merge
framework that accounts for the sparsity and redundancy of KV tokens across
different heads. Additionally, we implement the head-wise parallel compression
through a zero-class mechanism to enhance efficiency. Extensive experiments
demonstrate our SOTA performance even under extreme compression ratios. EMS
consistently achieves the lowest perplexity, improves scores by over 1.28
points across four LLMs on LongBench under a 256 cache budget, and preserves
95% retrieval accuracy with a cache budget less than 2% of the context length
in the Needle-in-a-Haystack task.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 16:35:13 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 15:29:03 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Li",
"Yingxin",
""
],
[
"Li",
"Ye",
""
],
[
"Meng",
"Yuan",
""
],
[
"Ma",
"Xinzhu",
""
],
[
"Geng",
"Zihan",
""
],
[
"Xia",
"Shutao",
""
],
[
"Wang",
"Zhi",
""
]
] |
2412.08620 | Halil Mutuk | Halil Mutuk | Revisiting light-flavor diquarks in the inverse matrix method of QCD sum
rules | 11 pages, 3 figures, 2 tables | Physical Review D 111, 034035 (2025) | 10.1103/PhysRevD.111.034035 | null | hep-ph hep-ex hep-lat | http://creativecommons.org/licenses/by/4.0/ | This study reexamines the spectroscopic parameters of light-flavor diquarks
within the framework of quantum chromodynamics sum rules (QCDSR) using the
inverse matrix method. Conventional QCDSR analyses are based on assumptions
such as quark-hadron duality and continuum models, which introduce a degree of
systematic uncertainty. The inverse matrix method circumvents these assumptions
by reformulating the problem as an inverse integral equation and expanding the
unknown spectral density using orthogonal Laguerre polynomials. This method
allows for a direct determination of spectral densities, thereby enhancing the
precision of predictions regarding resonance masses and decay constants. By
employing this methodology with regard to light-flavor diquarks ($sq$ and
$ud$), it is possible to extract the associated masses and decay constants. The
results indicate that the masses of diquarks with quantum numbers $J^P = 0^+$
and $J^P = 0^-$ are nearly degenerate. We compare our results regarding masses
and decay constants with those of other theoretical predictions, which could
prove a useful complementary tool in interpretation. Our results are consistent
with those in the literature and can be shown as evidence for the consistency
of the method. The results achieved in this study highlight the potential of
the inverse matrix method as a robust tool for exploring nonperturbative QCD
phenomena and elucidating the internal structure of exotic hadronic systems.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 18:40:32 GMT"
},
{
"version": "v2",
"created": "Thu, 12 Dec 2024 17:49:58 GMT"
},
{
"version": "v3",
"created": "Wed, 26 Feb 2025 20:00:06 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Mutuk",
"Halil",
""
]
] |
2412.08774 | Jungho Kim | Jungho Kim, Changwon Kang, Dongyoung Lee, Sehwan Choi, Jun Won Choi | ProtoOcc: Accurate, Efficient 3D Occupancy Prediction Using Dual Branch
Encoder-Prototype Query Decoder | Accepted to AAAI Conference on Artificial Intelligence 2025, 15
pages, 9 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we introduce ProtoOcc, a novel 3D occupancy prediction model
designed to predict the occupancy states and semantic classes of 3D voxels
through a deep semantic understanding of scenes. ProtoOcc consists of two main
components: the Dual Branch Encoder (DBE) and the Prototype Query Decoder
(PQD). The DBE produces a new 3D voxel representation by combining 3D voxel and
BEV representations across multiple scales through a dual branch structure.
This design enhances both performance and computational efficiency by providing
a large receptive field for the BEV representation while maintaining a smaller
receptive field for the voxel representation. The PQD introduces Prototype
Queries to accelerate the decoding process. Scene-Adaptive Prototypes are
derived from the 3D voxel features of input sample, while Scene-Agnostic
Prototypes are computed by applying Scene-Adaptive Prototypes to an Exponential
Moving Average during the training phase. By using these prototype-based
queries for decoding, we can directly predict 3D occupancy in a single step,
eliminating the need for iterative Transformer decoding. Additionally, we
propose the Robust Prototype Learning, which injects noise into prototype
generation process and trains the model to denoise during the training phase.
ProtoOcc achieves state-of-the-art performance with 45.02% mIoU on the
Occ3D-nuScenes benchmark. For single-frame method, it reaches 39.56% mIoU with
an inference speed of 12.83 FPS on an NVIDIA RTX 3090. Our code can be found at
https://github.com/SPA-junghokim/ProtoOcc.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 20:55:21 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:57:16 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Kim",
"Jungho",
""
],
[
"Kang",
"Changwon",
""
],
[
"Lee",
"Dongyoung",
""
],
[
"Choi",
"Sehwan",
""
],
[
"Choi",
"Jun Won",
""
]
] |
2412.08881 | Mohammed Hassan | Mohamed Sennary, Javier Rivera-Dean, Mohamed ElKabbash, Vladimir
Pervak, Maciej Lewenstein, Mohammed Th. Hassan | Ultrafast quantum light uncertainty dynamics in real time | 22 pages, 8 figures | null | null | null | physics.optics physics.app-ph quant-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Advancements in quantum optics and squeezed light generation have transformed
various domains of quantum science and technology. However, real-time quantum
dynamics remain an underexplored frontier. Here, we extend quantum optics into
the ultrafast regime, providing direct experimental evidence that quantum
uncertainty is not a static constraint but evolves dynamically with the system
state and interactions. Using ultrafast squeezed light generated via a
four-wave mixing nonlinear process, we observe the temporal dynamics of
amplitude uncertainty, demonstrating that quantum uncertainty is a controllable
and tunable physical quantity. This offers new insights into fundamental
quantum mechanics in real-time. Additionally, we demonstrate control over the
quantum state of light by switching between amplitude and phase squeezing. Our
ability to generate and manipulate ultrafast squeezed light waveforms with
attosecond resolution unlocks exciting possibilities for quantum technologies,
including petahertz scale secure quantum communication, quantum computing, and
ultrafast spectroscopy. We also introduce an ultrafast quantum encryption
protocol leveraging squeezed light for secure digital communication at
unprecedented speeds. This work paves the way for exploring quantum uncertainty
dynamics and establishes the foundation for the emerging field of ultrafast
quantum science.
| [
{
"version": "v1",
"created": "Thu, 12 Dec 2024 02:33:45 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 19:49:45 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Sennary",
"Mohamed",
""
],
[
"Rivera-Dean",
"Javier",
""
],
[
"ElKabbash",
"Mohamed",
""
],
[
"Pervak",
"Vladimir",
""
],
[
"Lewenstein",
"Maciej",
""
],
[
"Hassan",
"Mohammed Th.",
""
]
] |
2412.09149 | Nico Messikommer | Nico Messikommer, Jiaxu Xing, Elie Aljalbout, Davide Scaramuzza | Student-Informed Teacher Training | null | International Conference on Learning Representations (ICLR) 2025 | null | null | cs.RO cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Imitation learning with a privileged teacher has proven effective for
learning complex control behaviors from high-dimensional inputs, such as
images. In this framework, a teacher is trained with privileged task
information, while a student tries to predict the actions of the teacher with
more limited observations, e.g., in a robot navigation task, the teacher might
have access to distances to nearby obstacles, while the student only receives
visual observations of the scene. However, privileged imitation learning faces
a key challenge: the student might be unable to imitate the teacher's behavior
due to partial observability. This problem arises because the teacher is
trained without considering if the student is capable of imitating the learned
behavior. To address this teacher-student asymmetry, we propose a framework for
joint training of the teacher and student policies, encouraging the teacher to
learn behaviors that can be imitated by the student despite the latters'
limited access to information and its partial observability. Based on the
performance bound in imitation learning, we add (i) the approximated action
difference between teacher and student as a penalty term to the reward function
of the teacher, and (ii) a supervised teacher-student alignment step. We
motivate our method with a maze navigation task and demonstrate its
effectiveness on complex vision-based quadrotor flight and manipulation tasks.
| [
{
"version": "v1",
"created": "Thu, 12 Dec 2024 10:34:26 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 12:54:47 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Messikommer",
"Nico",
""
],
[
"Xing",
"Jiaxu",
""
],
[
"Aljalbout",
"Elie",
""
],
[
"Scaramuzza",
"Davide",
""
]
] |
2412.09756 | Rayne Holland Ph. D | Rayne Holland, Seyit Camtepe, Chandra Thapa, Jason Xue | Private Synthetic Data Generation in Small Memory | 24 Pages, 1 Table, 3 Figures, 3 Algorithms | null | null | null | cs.CR cs.DS | http://creativecommons.org/licenses/by/4.0/ | We propose $\mathtt{PrivHP}$, a lightweight synthetic data generator with
\textit{differential privacy} guarantees. $\mathtt{PrivHP}$ uses a novel
hierarchical decomposition that approximates the input's cumulative
distribution function (CDF) in bounded memory. It balances hierarchy depth,
noise addition, and pruning of low-frequency subdomains while preserving
frequent ones. Private sketches estimate subdomain frequencies efficiently
without full data access.
A key feature is the pruning parameter $k$, which controls the trade-off
between space and utility. We define the skew measure $\mathtt{tail}_k$,
capturing all but the top $k$ subdomain frequencies. Given a dataset
$\mathcal{X}$, $\mathtt{PrivHP}$ uses $M=\mathcal{O}(k\log^2 |X|)$ space and,
for input domain $\Omega = [0,1]$, ensures $\varepsilon$-differential privacy.
It yields a generator with expected Wasserstein distance: \[
\mathcal{O}\left(\frac{\log^2 M}{\varepsilon n} +
\frac{||\mathtt{tail}_k(\mathcal{X})||_1}{M n}\right) \] from the empirical
distribution. This parameterized trade-off offers a level of flexibility
unavailable in prior work. We also provide interpretable utility bounds that
account for hierarchy depth, privacy noise, pruning, and frequency estimation
errors.
| [
{
"version": "v1",
"created": "Thu, 12 Dec 2024 23:24:05 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 00:58:38 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Holland",
"Rayne",
""
],
[
"Camtepe",
"Seyit",
""
],
[
"Thapa",
"Chandra",
""
],
[
"Xue",
"Jason",
""
]
] |
2412.10193 | Yair Schiff | Yair Schiff, Subham Sekhar Sahoo, Hao Phung, Guanghan Wang, Sam
Boshar, Hugo Dalla-torre, Bernardo P. de Almeida, Alexander Rush, Thomas
Pierrot, Volodymyr Kuleshov | Simple Guidance Mechanisms for Discrete Diffusion Models | ICLR 2025; Code to reproduce our experiments is available here:
https://github.com/kuleshov-group/discrete-diffusion-guidance | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Diffusion models for continuous data gained widespread adoption owing to
their high quality generation and control mechanisms. However, controllable
diffusion on discrete data faces challenges given that continuous guidance
methods do not directly apply to discrete diffusion. Here, we provide a
straightforward derivation of classifier-free and classifier-based guidance for
discrete diffusion, as well as a new class of diffusion models that leverage
uniform noise and that are more guidable because they can continuously edit
their outputs. We improve the quality of these models with a novel
continuous-time variational lower bound that yields state-of-the-art
performance, especially in settings involving guidance or fast generation.
Empirically, we demonstrate that our guidance mechanisms combined with uniform
noise diffusion improve controllable generation relative to autoregressive and
diffusion baselines on several discrete data domains, including genomic
sequences, small molecule design, and discretized image generation.
| [
{
"version": "v1",
"created": "Fri, 13 Dec 2024 15:08:30 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 20:08:59 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Schiff",
"Yair",
""
],
[
"Sahoo",
"Subham Sekhar",
""
],
[
"Phung",
"Hao",
""
],
[
"Wang",
"Guanghan",
""
],
[
"Boshar",
"Sam",
""
],
[
"Dalla-torre",
"Hugo",
""
],
[
"de Almeida",
"Bernardo P.",
""
],
[
"Rush",
"Alexander",
""
],
[
"Pierrot",
"Thomas",
""
],
[
"Kuleshov",
"Volodymyr",
""
]
] |
2412.10739 | Jingyu Zhang | Jingyu Zhang, Yilei Wang, Lang Qian, Peng Sun, Zengwen Li, Sudong
Jiang, Maolin Liu, Liang Song | DSRC: Learning Density-insensitive and Semantic-aware Collaborative
Representation against Corruptions | Accepted by AAAI2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | As a potential application of Vehicle-to-Everything (V2X) communication,
multi-agent collaborative perception has achieved significant success in 3D
object detection. While these methods have demonstrated impressive results on
standard benchmarks, the robustness of such approaches in the face of complex
real-world environments requires additional verification. To bridge this gap,
we introduce the first comprehensive benchmark designed to evaluate the
robustness of collaborative perception methods in the presence of natural
corruptions typical of real-world environments. Furthermore, we propose DSRC, a
robustness-enhanced collaborative perception method aiming to learn
Density-insensitive and Semantic-aware collaborative Representation against
Corruptions. DSRC consists of two key designs: i) a semantic-guided
sparse-to-dense distillation framework, which constructs multi-view dense
objects painted by ground truth bounding boxes to effectively learn
density-insensitive and semantic-aware collaborative representation; ii) a
feature-to-point cloud reconstruction approach to better fuse critical
collaborative representation across agents. To thoroughly evaluate DSRC, we
conduct extensive experiments on real-world and simulated datasets. The results
demonstrate that our method outperforms SOTA collaborative perception methods
in both clean and corrupted conditions. Code is available at
https://github.com/Terry9a/DSRC.
| [
{
"version": "v1",
"created": "Sat, 14 Dec 2024 08:19:30 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 13:17:17 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zhang",
"Jingyu",
""
],
[
"Wang",
"Yilei",
""
],
[
"Qian",
"Lang",
""
],
[
"Sun",
"Peng",
""
],
[
"Li",
"Zengwen",
""
],
[
"Jiang",
"Sudong",
""
],
[
"Liu",
"Maolin",
""
],
[
"Song",
"Liang",
""
]
] |
2412.10931 | Michael Galperin | Haoran Sun, Upendra Harbola, Shaul Mukamel, Michael Galperin | Two-dimensional spectroscopy of open quantum systems | 10 pages, 6 figures | J. Phys. Chem. Lett. 16, 2008-2015 (2025) | 10.1021/acs.jpclett.4c03597 | null | cond-mat.mes-hall physics.chem-ph physics.optics | http://creativecommons.org/licenses/by/4.0/ | Two-dimensional spectroscopy is discussed for open quantum systems with
multiple simultaneously measurable fluxes. In particular, we discuss a junction
where optical measurements of photon flux are complemented with simultaneous
transport measurements of electron currents. Theory of two-dimensional
spectroscopy in both fluxes is developed employing non-self-consistent
nonequilibrium Green's function formulation. Theoretical derivations are
illustrated with numerical simulations within generic junction model.
| [
{
"version": "v1",
"created": "Sat, 14 Dec 2024 18:55:54 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Sun",
"Haoran",
""
],
[
"Harbola",
"Upendra",
""
],
[
"Mukamel",
"Shaul",
""
],
[
"Galperin",
"Michael",
""
]
] |
2412.11758 | Gabriel De Jesus | Gabriel de Jesus and S\'ergio Nunes | Establishing a Foundation for Tetun Text Ad-Hoc Retrieval: Stemming,
Indexing, Retrieval, and Ranking | Refine the title and improve the content (version 2) | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Searching for information on the internet and digital platforms to satisfy an
information need requires effective retrieval solutions. However, such
solutions are not yet available for Tetun, making it challenging to find
relevant documents for text-based search queries in this language. To address
these challenges, this study investigates Tetun text retrieval with a focus on
the ad-hoc retrieval task. It begins by developing essential language resources
-- including a list of stopwords, a stemmer, and a test collection -- which
serve as foundational components for solutions tailored to Tetun text
retrieval. Various strategies are then explored using both document titles and
content to evaluate retrieval effectiveness. The results show that retrieving
document titles, after removing hyphens and apostrophes without applying
stemming, significantly improves retrieval performance compared to the
baseline. Efficiency increases by 31.37%, while effectiveness achieves an
average gain of 9.40% in MAP@10 and 30.35% in nDCG@10 with DFR BM25. Beyond the
top-10 cutoff point, Hiemstra LM demonstrates strong performance across various
retrieval strategies and evaluation metrics. Contributions of this work include
the development of Labadain-Stopwords (a list of 160 Tetun stopwords),
Labadain-Stemmer (a Tetun stemmer with three variants), and
Labadain-Avaliad\'or (a Tetun test collection containing 59 topics, 33,550
documents, and 5,900 qrels).
| [
{
"version": "v1",
"created": "Mon, 16 Dec 2024 13:22:34 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 09:02:20 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"de Jesus",
"Gabriel",
""
],
[
"Nunes",
"Sérgio",
""
]
] |
2412.12015 | Joel Swallow | NA62 Collaboration | Observation of the $K^{+}\rightarrow\pi^{+}\nu\bar{\nu}$ decay and
measurement of its branching ratio | null | JHEP 02 (2025) 191 | 10.1007/JHEP02(2025)191 | CERN-EP-2024-343 | hep-ex | http://creativecommons.org/licenses/by/4.0/ | A measurement of the $K^{+}\rightarrow\pi^{+}\nu\bar{\nu}$ decay by the NA62
experiment at the CERN SPS is presented, using data collected in 2021 and 2022.
This dataset was recorded, after modifications to the beamline and detectors,
at a higher instantaneous beam intensity with respect to the 2016--2018 data
taking. Combining NA62 data collected in 2016--2022, a measurement of
$B(K^{+}\rightarrow\pi^{+}\nu\bar{\nu}) = \left( 13.0^{+ 3.3}_{- 3.0}
\right)\times10^{-11}$ is reported. With $51$ signal candidates observed and an
expected background of $18^{+3}_{-2}$ events,
$B(K^{+}\rightarrow\pi^{+}\nu\bar{\nu})$ becomes the smallest branching ratio
measured with a signal significance above $5\,\sigma$.
| [
{
"version": "v1",
"created": "Mon, 16 Dec 2024 17:40:50 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 13:23:31 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"NA62 Collaboration",
"",
""
]
] |
2412.12255 | Yuan Xin | A. Liam Fitzpatrick, Emanuel Katz, Yuan Xin | Toolkit for General 2d Scalar Potential in LCT | 30 pages, 11 figures, v2: minor fixes | null | null | null | hep-th cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | We present efficient algorithms for obtaining the Hamiltonian in Lightcone
Conformal Truncation (LCT) for a 2d scalar field with a generic potential. We
apply this method to the sine-Gordon and sinh-Gordon models in $1+1d$, and find
precise agreement with integrability results when the scaling dimension
$\Delta$ of the deforming cosine/cosinh potential is in the range $ \Delta \leq
1$. The agreement provides additional evidence for a recent conjecture for how
to compute the effective lightcone Hamiltonian in this class of models. In
addition, to high precision, we provide the first direct confirmation for the
conjectured self-duality of the sinh-Gordon model ($\Delta < 0)$, which relates
$\Delta \leftrightarrow 4/\Delta$. As the dimension approaches the upper limit
$\Delta=1$ from below, we show analytically that the Hamiltonian matrix
elements exactly reproduce those of a free Majorana fermion, demonstrating how
bosonization is manifested in the LCT basis. We comment on the possible
extension of the approach to $\Delta > 1$.
| [
{
"version": "v1",
"created": "Mon, 16 Dec 2024 19:00:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 18:46:52 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Fitzpatrick",
"A. Liam",
""
],
[
"Katz",
"Emanuel",
""
],
[
"Xin",
"Yuan",
""
]
] |
2412.13894 | Antony Lewis | Antony Lewis and Ewan Chamberlain | Understanding acoustic scale observations: the one-sided fight against
$\Lambda$ | 11 pages, 7 Figures. v2: simplified prior ranges/table; added Fig 7
to show $\Lambda$CDM dependence. Misc clarifications and references | null | null | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | The cosmic microwave background (CMB) and baryon acoustic oscillations (BAO)
provide precise measurements of the cosmic expansion history through the
comoving acoustic scale. The CMB angular scale measurement $\theta_*$ is
particularly robust, constraining the ratio of the sound horizon to the angular
diameter distance to last scattering independently of the late-time
cosmological model. For models with standard early-universe physics, this
measurement strongly constrains possible deviations from $\Lambda$CDM at late
times. We show that the null energy condition imposes strict inequalities on
the BAO observables $D_H(z)$, $D_M(z)$, $D_V(z)$ and $F_{\rm AP}(z)$ relative
to $\Lambda$CDM predictions. These inequalities demonstrate that certain
deviations from $\Lambda$CDM are impossible for any physical non-interacting
dark energy model that respects the null energy condition. We also identify the
regions of parameter space in the CPL parameterization $w(a) = w_0 + w_a(1-a)$
that can give predictions consistent with both the null energy condition and
the observed CMB scale. While current DESI DR1 BAO measurements exhibit slight
joint-constraint parameter tensions with $\Lambda$CDM, this tension only arises
in directions that are inconsistent with the null-energy condition, so
$\Lambda$CDM is favoured by current acoustic scale measurements unless the
null-energy condition is violated.
| [
{
"version": "v1",
"created": "Wed, 18 Dec 2024 14:35:07 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 10:16:25 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Lewis",
"Antony",
""
],
[
"Chamberlain",
"Ewan",
""
]
] |
2412.13993 | John M. Hanna | John M. Hanna, Hugues Talbot, Irene E. Vignon-Clementel | Improved Physics-informed neural networks loss function regularization
with a variance-based term | null | null | null | null | math.OC cs.LG | http://creativecommons.org/licenses/by/4.0/ | In machine learning and statistical modeling, the mean square or absolute
error is commonly used as an error metric, also called a "loss function." While
effective in reducing the average error, this approach may fail to address
localized outliers, leading to significant inaccuracies in regions with sharp
gradients or discontinuities. This issue is particularly evident in
physics-informed neural networks (PINNs), where such localized errors are
expected and affect the overall solution. To overcome this limitation, we
propose a novel loss function that combines the mean and the standard deviation
of the chosen error metric. By minimizing this combined loss function, the
method ensures a more uniform error distribution and reduces the impact of
localized high-error regions. The proposed loss function is easy to implement
and tested on problems of varying complexity: the 1D Poisson equation, the
unsteady Burgers' equation, 2D linear elastic solid mechanics, and 2D steady
Navier-Stokes equations. Results demonstrate improved solution quality and
lower maximum error compared to the standard mean-based loss, with minimal
impact on computational time.
| [
{
"version": "v1",
"created": "Wed, 18 Dec 2024 16:11:45 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 20:31:35 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Hanna",
"John M.",
""
],
[
"Talbot",
"Hugues",
""
],
[
"Vignon-Clementel",
"Irene E.",
""
]
] |
2412.14340 | Alexis Fox | Alexis Fox, Samarth Swarup, Abhijin Adiga | A Unifying Information-theoretic Perspective on Evaluating Generative
Models | null | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Considering the difficulty of interpreting generative model output, there is
significant current research focused on determining meaningful evaluation
metrics. Several recent approaches utilize "precision" and "recall," borrowed
from the classification domain, to individually quantify the output fidelity
(realism) and output diversity (representation of the real data variation),
respectively. With the increase in metric proposals, there is a need for a
unifying perspective, allowing for easier comparison and clearer explanation of
their benefits and drawbacks. To this end, we unify a class of
kth-nearest-neighbors (kNN)-based metrics under an information-theoretic lens
using approaches from kNN density estimation. Additionally, we propose a
tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall
Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity
and two distinct aspects of diversity, inter- and intra-class. Our
domain-agnostic metric, derived from the information-theoretic concepts of
entropy and cross-entropy, can be dissected for both sample- and mode-level
analysis. Our detailed experimental results demonstrate the sensitivity of our
metric components to their respective qualities and reveal undesirable
behaviors of other metrics.
| [
{
"version": "v1",
"created": "Wed, 18 Dec 2024 21:17:02 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Jan 2025 00:02:00 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 18:25:41 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Fox",
"Alexis",
""
],
[
"Swarup",
"Samarth",
""
],
[
"Adiga",
"Abhijin",
""
]
] |
2412.14434 | Nathan Seiberg | Nathan Seiberg | Anomalous Continuous Translations | 30 pages, minor changes | null | null | null | cond-mat.str-el hep-th math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We discuss a large class of non-relativistic continuum field theories where
the Euclidean spatial symmetry of the classical theory is violated in the
quantum theory by an Adler-Bell-Jackiw-like anomaly. In particular, the
continuous translation symmetry of the classical theory is broken in the
quantum theory to a discrete symmetry. Furthermore, that discrete symmetry is
extended by an internal symmetry, making it non-Abelian. This presentation
streamlines and extends the discussion in [1]. In an Appendix, we present an
elementary introduction to 't Hooft and Adler-Bell-Jackiw anomalies using a
well-known system.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 01:32:15 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:31:27 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Seiberg",
"Nathan",
""
]
] |
2412.14488 | Chuan He | Chuan He | A stochastic first-order method with multi-extrapolated momentum for
highly smooth unconstrained optimization | Some claims about the contributions were overstated, and we are
currently working on the revision | null | null | null | math.OC cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | In this paper, we consider an unconstrained stochastic optimization problem
where the objective function exhibits high-order smoothness. Specifically, we
propose a new stochastic first-order method (SFOM) with multi-extrapolated
momentum, in which multiple extrapolations are performed in each iteration,
followed by a momentum update based on these extrapolations. We demonstrate
that the proposed SFOM can accelerate optimization by exploiting the high-order
smoothness of the objective function $f$. Assuming that the $p$th-order
derivative of $f$ is Lipschitz continuous for some $p\ge2$, and under
additional mild assumptions, we establish that our method achieves a sample
complexity of $\widetilde{\mathcal{O}}(\epsilon^{-(3p+1)/p})$ for finding a
point $x$ such that $\mathbb{E}[\|\nabla f(x)\|]\le\epsilon$. To the best of
our knowledge, this is the first SFOM to leverage arbitrary-order smoothness of
the objective function for acceleration, resulting in a sample complexity that
improves upon the best-known results without assuming the mean-squared
smoothness condition. Preliminary numerical experiments validate the practical
performance of our method and support our theoretical findings.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 03:22:47 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Jan 2025 13:52:14 GMT"
},
{
"version": "v3",
"created": "Wed, 26 Feb 2025 22:28:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"He",
"Chuan",
""
]
] |
2412.14711 | Ziteng Wang | Ziteng Wang, Jun Zhu, Jianfei Chen | ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing | ICLR 2025 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Sparsely activated Mixture-of-Experts (MoE) models are widely adopted to
scale up model capacity without increasing the computation budget. However,
vanilla TopK routers are trained in a discontinuous, non-differentiable way,
limiting their performance and scalability. To address this issue, we propose
ReMoE, a fully differentiable MoE architecture that offers a simple yet
effective drop-in replacement for the conventional TopK+Softmax routing,
utilizing ReLU as the router instead. We further propose methods to regulate
the router's sparsity while balancing the load among experts. ReMoE's
continuous nature enables efficient dynamic allocation of computation across
tokens and layers, while also exhibiting domain specialization. Our experiments
demonstrate that ReMoE consistently outperforms vanilla TopK-routed MoE across
various model sizes, expert counts, and levels of granularity. Furthermore,
ReMoE exhibits superior scalability with respect to the number of experts,
surpassing traditional MoE architectures. The implementation based on
Megatron-LM is available at https://github.com/thu-ml/ReMoE.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 10:21:20 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:33:09 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Wang",
"Ziteng",
""
],
[
"Zhu",
"Jun",
""
],
[
"Chen",
"Jianfei",
""
]
] |
2412.15291 | Chenxiao Yu | Chenxiao Yu, Jinyi Ye, Yuangang Li, Zhaotian Weng, Zheng Li, Emilio
Ferrara, Xiyang Hu, Yue Zhao | A Large-Scale Simulation on Large Language Models for Decision-Making in
Political Science | arXiv admin note: substantial text overlap with arXiv:2411.03321 This
version adds a new model to our experimental setup, modifies the paper's main
discussion, and updates the authorship list | null | null | null | cs.CL cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While LLMs have demonstrated remarkable capabilities in text generation and
reasoning, their ability to simulate human decision-making -- particularly in
political contexts -- remains an open question. However, modeling voter
behavior presents unique challenges due to limited voter-level data, evolving
political landscapes, and the complexity of human reasoning. In this study, we
develop a theory-driven, multi-step reasoning framework that integrates
demographic, temporal and ideological factors to simulate voter decision-making
at scale. Using synthetic personas calibrated to real-world voter data, we
conduct large-scale simulations of recent U.S. presidential elections. Our
method significantly improves simulation accuracy while mitigating model
biases. We examine its robustness by comparing performance across different
LLMs. We further investigate the challenges and constraints that arise from
LLM-based political simulations. Our work provides both a scalable framework
for modeling political decision-making behavior and insights into the promise
and limitations of using LLMs in political science research.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 07:10:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:47:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Yu",
"Chenxiao",
""
],
[
"Ye",
"Jinyi",
""
],
[
"Li",
"Yuangang",
""
],
[
"Weng",
"Zhaotian",
""
],
[
"Li",
"Zheng",
""
],
[
"Ferrara",
"Emilio",
""
],
[
"Hu",
"Xiyang",
""
],
[
"Zhao",
"Yue",
""
]
] |
2412.15399 | Shokoufe Faraji | Daniel Amaro, Shokoufe Faraji | Vacuum polarization effects in the background of a deformed compact
object and implications for photon velocity | null | Physical Review D, Vol. 111, No. 4, 5 February 2025 | 10.1103/PhysRevD.111.045003 | null | gr-qc hep-th math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies the impact of vacuum polarization on light propagation in
the background of a distorted, deformed compact object. Focusing on a spacetime
containing two quadrupole parameters associated with the central object and
external fields, we explore how these parameters influence observable effects
as dynamical degrees of freedom. In this setup, we investigate electromagnetic
birefringence, noting distinct polarization-dependent photon velocity
variations and gravitational lens effects. Although current resolution may
limit detection, future high-precision observations could reveal these quantum
electrodynamics (QED) induced birefringence effects, advancing our
understanding of vacuum birefringence in astrophysical contexts. We further
analyze the dependence of shadow properties on the model's variables, using
observational data from Sgr A*.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 21:04:07 GMT"
},
{
"version": "v2",
"created": "Thu, 6 Feb 2025 03:16:25 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Amaro",
"Daniel",
""
],
[
"Faraji",
"Shokoufe",
""
]
] |
2412.16097 | Matteo Nerini | Matteo Nerini, Bruno Clerckx | Dual-Polarized Beyond Diagonal RIS | Accepted by IEEE for publication | null | null | null | cs.IT eess.SP math.IT | http://creativecommons.org/licenses/by/4.0/ | Beyond diagonal reconfigurable intelligent surface (BD-RIS) is a family of
RIS architectures more flexible than conventional RIS. While BD-RIS has been
primarily analyzed assuming uni-polarized systems, modern wireless deployments
are dual-polarized. To address this gap, this paper investigates the
fundamental limits of dual-polarized BD-RIS-aided systems. We derive the
scaling laws governing the performance of BD-RIS and the Pareto frontier of the
trade-off between performance and circuit complexity enabled by BD-RIS.
Theoretical results show that the group-connected RIS with group size 2
provides remarkable gains over conventional RIS in both Rayleigh and
line-of-sight (LoS) channels, while maintaining a reduced circuit complexity.
| [
{
"version": "v1",
"created": "Fri, 20 Dec 2024 17:41:02 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 17:43:42 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Nerini",
"Matteo",
""
],
[
"Clerckx",
"Bruno",
""
]
] |
2412.16382 | Amin Bigdeli | Amin Bigdeli, Negar Arabzadeh, Ebrahim Bagheri, Charles L. A. Clarke | EMPRA: Embedding Perturbation Rank Attack against Neural Ranking Models | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent research has shown that neural information retrieval techniques may be
susceptible to adversarial attacks. Adversarial attacks seek to manipulate the
ranking of documents, with the intention of exposing users to targeted content.
In this paper, we introduce the Embedding Perturbation Rank Attack (EMPRA)
method, a novel approach designed to perform adversarial attacks on black-box
Neural Ranking Models (NRMs). EMPRA manipulates sentence-level embeddings,
guiding them towards pertinent context related to the query while preserving
semantic integrity. This process generates adversarial texts that seamlessly
integrate with the original content and remain imperceptible to humans. Our
extensive evaluation conducted on the widely-used MS MARCO V1 passage
collection demonstrate the effectiveness of EMPRA against a wide range of
state-of-the-art baselines in promoting a specific set of target documents
within a given ranked results. Specifically, EMPRA successfully achieves a
re-ranking of almost 96% of target documents originally ranked between 51-100
to rank within the top 10. Furthermore, EMPRA does not depend on surrogate
models for adversarial text generation, enhancing its robustness against
different NRMs in realistic settings.
| [
{
"version": "v1",
"created": "Fri, 20 Dec 2024 22:36:19 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 23:46:31 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Bigdeli",
"Amin",
""
],
[
"Arabzadeh",
"Negar",
""
],
[
"Bagheri",
"Ebrahim",
""
],
[
"Clarke",
"Charles L. A.",
""
]
] |
2412.16795 | Edward Witten | Edward Witten | Introduction to Black Hole Thermodynamics | 127 pp, 30 figures; added references and minor corrections in v. 2,
v. 3 | null | null | null | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | These notes aim to provide an introduction to the basics of black hole
thermodynamics. After explaining Bekenstein's original proposal that black
holes have entropy, we discuss Hawking's discovery of black hole radiation, its
analog for Rindler space in the Unruh effect, the Euclidean approach to black
hole thermodynamics, some basics about von Neumann entropy and its
applications, the Ryu-Takayanagi formula, and the nature of a white hole.
| [
{
"version": "v1",
"created": "Sat, 21 Dec 2024 22:44:38 GMT"
},
{
"version": "v2",
"created": "Sun, 29 Dec 2024 21:27:10 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 18:13:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Witten",
"Edward",
""
]
] |
2412.17569 | Jaume Llabr\'es | Jaume Llabr\'es, Sara Oliver-Bonafoux, Celia Anteneodo, Ra\'ul Toral | Complete aging in the noisy voter model enhances consensus formation | null | Complete aging in the noisy voter model enhances consensus
formation. Llabr\'es J, Oliver-Bonafoux S, Anteneodo C, Toral, R. Chaos,
Solitons & Fractals 194, 116153 (2025) | 10.1016/j.chaos.2025.116153 | null | physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | We investigate the effects of aging in the noisy voter model considering that
the probability to change states decays algebraically with age $\tau$, defined
as the time elapsed since adopting the current state. We study the complete
aging scenario, which incorporates aging to both mechanisms of interaction:
herding and idiosyncratic behavior, and compare it with the partial aging case,
where aging affects only the herding mechanism. Analytical mean-field equations
are derived, finding excellent agreement with agent-based simulations on a
complete graph. We observe that complete aging enhances consensus formation,
shifting the critical point to higher values compared to the partial aging
case. However, when the aging probability decays asymptotically to zero for
large $\tau$, a steady state is not always attained for complete aging.
| [
{
"version": "v1",
"created": "Mon, 23 Dec 2024 13:42:37 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Llabrés",
"Jaume",
""
],
[
"Oliver-Bonafoux",
"Sara",
""
],
[
"Anteneodo",
"Celia",
""
],
[
"Toral",
"Raúl",
""
]
] |
2412.18556 | Vishal Singh | Vishal Singh, Theshani Nuradha, and Mark M. Wilde | Extendible quantum measurements and limitations on classical
communication | 6+12 pages. Submission to the 2025 International Symposium on
Information Theory to be held at University of Michigan. v2: Added Appendix
D, demonstrating the efficient computability of the SDP upper bound on the
n-shot classical capacity of a channel for a fixed 'k' | null | null | null | quant-ph cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unextendibility of quantum states and channels is inextricably linked to the
no-cloning theorem of quantum mechanics, it has played an important role in
understanding and quantifying entanglement, and more recently it has found
applications in providing limitations on quantum error correction and
entanglement distillation. Here we generalize the framework of unextendibility
to quantum measurements and define $k$-extendible measurements for every
integer $k\ge 2$. Our definition provides a hierarchy of semidefinite
constraints that specify a set of measurements containing every measurement
that can be realized by local operations and one-way classical communication.
Furthermore, the set of $k$-extendible measurements converges to the set of
measurements that can be realized by local operations and one-way classical
communication as $k\to \infty$. To illustrate the utility of $k$-extendible
measurements, we establish a semidefinite programming upper bound on the
one-shot classical capacity of a channel, which outperforms the best known
efficiently computable bound from [Matthews and Wehner, IEEE Trans. Inf. Theory
60, pp. 7317-7329 (2014)] and also leads to efficiently computable upper bounds
on the $n$-shot classical capacity of a channel.
| [
{
"version": "v1",
"created": "Tue, 24 Dec 2024 17:12:45 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 21:55:48 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Singh",
"Vishal",
""
],
[
"Nuradha",
"Theshani",
""
],
[
"Wilde",
"Mark M.",
""
]
] |
2412.18625 | Xiaodong Zhang | Xiaodong Zhang | Solving Crystal Structures by Carrying Out the Calculation of the
Single-Atom R1 Method in a Lottery Mode | 22 pages, 9 figures | null | null | null | physics.chem-ph physics.comp-ph physics.data-an | http://creativecommons.org/licenses/by/4.0/ | As originally designed [Zhang & Donahue (2024), Acta Cryst. A80, 2370248.],
after one cycle of calculation, the single-atom R1 (sR1) method required a user
to intelligently determine a partial structure to start the next cycle. In this
paper, a lottery scheme has been designed to randomly split a parent model into
two child models. This allows the calculation to be carried out in care-free
manner. By chance, one child model may have higher amounts of "good" atoms than
the parent model. Thus, its expansion in the next cycle favors an improved
model. These "lucky" results are carried onto the next cycles. while "unlucky"
results in which no improvements occur are discarded. Furthermore, unchanged
models are carried onto the next cycles in those "unlucky" occasions. On
average a child model has the same fraction of "good" atoms as the parent. Only
a substantial statistical fluctuation results in appreciable deviation. This
lottery scheme works because such fluctuations do happen. Indeed, test
applications with the computing power accessibly by the author have
demonstrated that the designed scheme can drive an sR1 calculation to or close
to reaching a correct structure solution.
| [
{
"version": "v1",
"created": "Thu, 19 Dec 2024 12:52:33 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 21:28:15 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Zhang",
"Xiaodong",
""
]
] |
2412.19528 | Zipeng Wang | Chuhan Sun and Zipeng Wang | On the end-point of Stein-Weiss inequality | We improve the result from the previous version to include the case
q>1 | null | null | null | math.CA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper has two purposes. First, we show that the classical Stein-Weiss
inequality is true for p=1. Second, by considering a family of strong
fractional integral operators whose kernels have singularity on every
coordinate subspace, we extend this end-point result to the multi-parameter
setting.
| [
{
"version": "v1",
"created": "Fri, 27 Dec 2024 08:42:32 GMT"
},
{
"version": "v2",
"created": "Sat, 18 Jan 2025 14:01:30 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 09:56:18 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Sun",
"Chuhan",
""
],
[
"Wang",
"Zipeng",
""
]
] |
2412.19704 | Bhupendra Chauhan | B. Chauhan, R. Tripathi | Re-parameterization Invariance of FRW Model: Supervariable and BRST
Approaches | LaTeX file, 28 pages, version to appear in NPB. arXiv admin note:
text overlap with arXiv:1912.12909 | B. Cahuhan, R. Tripathi, Nucl. Phys. B 1012, 116816 (2025) | null | null | hep-th | http://creativecommons.org/licenses/by/4.0/ | We perform the Becchi-Rouet-Stora-Tyutin (BRST) quantization of a $(0 +
1)$-dimensional cosmological Friedmann-Robertson-Walker (FRW) model. This
quantization leverages the classical infinitesimal and continuous
re-parameterization symmetry transformations of the system. To derive the
nilpotent re-parameterization invariant BRST-anti-BRST symmetry transformations
for the scale factor and corresponding momentum variables present in the
cosmological FRW model, we employ the modified Bonora-Tonin supervariable
approach (MBTSA) to BRST formalism. Through this approach, we also establish
the BRST-anti-BRST invariant Curci-Ferrari (CF)-type restriction for this
cosmological re-parameterization invariant model. Further, we obtain the
off-shell nilpotent quantum BRST-anti-BRST symmetry transformations for other
variables within the model using the (anti-)chiral supervariable approach
(ACSA) to BRST formalism. Within the framework of ACSA, the CF-type restriction
is demonstrated through two key aspects: $(i)$ the invariance of the coupled
Lagrangians under symmetry transformations, and $(ii)$ the absolute
anti-commutativity of the conserved BRST-anti-BRST charges. Notably, applying
the MBTSA to a physical cosmological system, specifically a one-dimensional
one, constitutes a novel contribution to this work. Additionally, in the
application of ACSA, we restrict our analysis to (anti-)chiral super expansions
of supervariables, leading to the unique observation of the absolute
anti-commutativity of the conserved BRST-anti-BRST charges. Moreover, we
highlight that the CF-type restriction demonstrates a universal nature,
remaining consistent across any re-parameterization invariant models in general
D-dimensional spacetime.
| [
{
"version": "v1",
"created": "Fri, 27 Dec 2024 15:58:29 GMT"
},
{
"version": "v2",
"created": "Wed, 1 Jan 2025 06:23:56 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 10:20:32 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Chauhan",
"B.",
""
],
[
"Tripathi",
"R.",
""
]
] |
2412.19902 | Vladislav Korenivski | D. Popadiuk, A. Vovk, S.A. Bunyaev, G.N. Kakazei, J.P. Araujo, P.
Strichovanec, P.A. Algarabel, V. Golub, A. Kravets, V. Korenivski, A.
Trzaskowska | Spin waves in Co$_2$FeGe films | 13 pages, 6 figures | null | null | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | The dynamic magnetic properties of Full Heusler alloy thin films of
Co$_2$FeGe, grown on MgO (001) substrates under different thermal conditions,
were investigated. Brillouin light scattering and ferromagnetic resonance
measurements revealed that depositing at room temperature followed by annealing
at 300 deg C for 1 hour produces the best results for maximizing magnetization,
exchange stiffness, and minimizing spin-dynamic dissipation in the films, which
are desirable characteristics for high-speed spintronic devices. Additionally,
strong hybridization of spin waves in the Damon-Eshbach geometry was observed,
which is attractive for applications in magnonic signal processing circuits.
| [
{
"version": "v1",
"created": "Fri, 27 Dec 2024 19:26:04 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 12:47:23 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Popadiuk",
"D.",
""
],
[
"Vovk",
"A.",
""
],
[
"Bunyaev",
"S. A.",
""
],
[
"Kakazei",
"G. N.",
""
],
[
"Araujo",
"J. P.",
""
],
[
"Strichovanec",
"P.",
""
],
[
"Algarabel",
"P. A.",
""
],
[
"Golub",
"V.",
""
],
[
"Kravets",
"A.",
""
],
[
"Korenivski",
"V.",
""
],
[
"Trzaskowska",
"A.",
""
]
] |
2412.20067 | Arif Er | Arif Er, Meng-Chwan Tan | Topological Gauge Theories with Sixteen Supercharges: Higher
$A_\infty$-categorification of Floer Homologies | 90 pp. Sect. 6 on a 3-category, and a generalization of Cao's
conjecture in Sect. 7, are new. This work is a sequel to arXiv:2410.18575,
and a third and final installment of the program initiated in
arXiv:2311.18302 | null | null | null | hep-th math.AG math.DG math.GT math.SG | http://creativecommons.org/licenses/by/4.0/ | This work is a sequel to [arXiv:2410.18575], and a third and final
installment of the program initiated in [arXiv:2311.18302]. We show how, via a
3d gauged Landau-Ginzburg model interpretation of certain topologically-twisted
5d $\mathcal{N} = 2$ and 8d $\mathcal{N} = 1$ gauge theories, one can derive
novel Fueter type $A_{\infty}$-2-categories that 2-categorify the
3d-Haydys-Witten, Haydys-Witten, and holomorphic Donaldson-Thomas Floer
homology of two, four, and five-manifolds, respectively. Via a 2d gauged
Landau-Ginzburg model interpretation of the aforementioned twisted gauge
theories, these Fueter type $A_{\infty}$-2-categories can be shown to be
equivalent to corresponding Fukaya-Seidel type $A_{\infty}$-categories. In the
8d case, one can also derive higher $A_{\infty}$-categories, such as a novel
Cauchy-Riemann-Fueter type $A_{\infty}$-3-category that 3-categorifies the
Haydys-Witten Floer homology of four-manifolds via a 4d gauged Landau-Ginzburg
model interpretation of the theory. Together with previous results from
[arXiv:2410.18575] and [arXiv:2311.18302], our work furnishes purely physical
proofs and generalizations of the mathematical conjectures by Bousseau [3],
Doan-Rezchikov [4], and Cao [5].
| [
{
"version": "v1",
"created": "Sat, 28 Dec 2024 07:42:29 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 03:22:12 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Er",
"Arif",
""
],
[
"Tan",
"Meng-Chwan",
""
]
] |
2412.20287 | Arghya Das | Pritha Dolai and Arghya Das | Boundary layers, transport and universal distribution in boundary driven
active systems | 31 pages, 10 figures | null | null | null | cond-mat.stat-mech cond-mat.soft | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We discuss analytical results for a run-and-tumble particle (RTP) in one
dimension in presence of boundary reservoirs. It exhibits `kinetic boundary
layers', nonmonotonous distribution, violation of Fourier's law, diffusion
facilitated current reversal and optimisation on tuning dynamical parameters,
and a Seebeck-like effect in the steady state. The spatial and internal degrees
of freedom together possess a symmetry, using which we find the eigenspectrum
for large systems. The eigenvalues are arranged in two bands which can mix in
certain conditions resulting in a crossover in the relaxation. The late time
distribution for large systems is obtained analytically; it retains a strong
and often dominant `active' contribution in the bulk rendering an effective
passive-like description inadequate. A nontrivial `Milne length' also emerges
in the dynamics. Finally, a novel universality is proposed in the absorbing
boundary problem for dynamics with short-ranged colored noise. Active particles
driven by active reservoirs may thus provide a common physical ground for
diverse and new nonequilibrium phenomena.
| [
{
"version": "v1",
"created": "Sat, 28 Dec 2024 21:50:11 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 11:06:48 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Dolai",
"Pritha",
""
],
[
"Das",
"Arghya",
""
]
] |
2412.20644 | Wonho Bae | Wonho Bae, Gabriel L. Oliveira, Danica J. Sutherland | Uncertainty Herding: One Active Learning Method for All Label Budgets | Accepted to ICLR2025 | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most active learning research has focused on methods which perform well when
many labels are available, but can be dramatically worse than random selection
when label budgets are small. Other methods have focused on the low-budget
regime, but do poorly as label budgets increase. As the line between "low" and
"high" budgets varies by problem, this is a serious issue in practice. We
propose uncertainty coverage, an objective which generalizes a variety of low-
and high-budget objectives, as well as natural, hyperparameter-light methods to
smoothly interpolate between low- and high-budget regimes. We call greedy
optimization of the estimate Uncertainty Herding; this simple method is
computationally fast, and we prove that it nearly optimizes the
distribution-level coverage. In experimental validation across a variety of
active learning tasks, our proposal matches or beats state-of-the-art
performance in essentially all cases; it is the only method of which we are
aware that reliably works well in both low- and high-budget settings.
| [
{
"version": "v1",
"created": "Mon, 30 Dec 2024 01:33:42 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 17:33:30 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Bae",
"Wonho",
""
],
[
"Oliveira",
"Gabriel L.",
""
],
[
"Sutherland",
"Danica J.",
""
]
] |
2412.21113 | Jonas Berx | Jonas Berx, Prashant Singh, Karel Proesmans | Positional information trade-offs in boundary-driven reaction-diffusion
systems | 13 pages, 9 figures | New J. Phys. 27, 023034 (2025) | 10.1088/1367-2630/adb7fc | null | cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | Individual components such as cells, particles, or agents within a larger
system often require detailed understanding of their relative position to act
accordingly, enabling the system as a whole to function in an organised and
efficient manner. Through the concept of positional information, such
components are able to specify their position in order to, e.g., create robust
spatial patterns or coordinate specific functionality. Such complex behaviour
generally occurs far from thermodynamic equilibrium and thus requires the
dissipation of free energy to sustain functionality. We show that in
boundary-driven simple exclusion systems with position-dependent Langmuir
kinetics, non-trivial Pareto-optimal trade-offs exist between the positional
information, rescaled entropy production rate and global reaction current.
Phase transitions in the optimal protocols that tune the densities of the
system boundaries emerge as a result, showing that distinct protocols are able
to exchange global optimality similar to phase coexistence in liquid-gas phase
transitions, and that increasing the positional information can lead to
diminishing returns when considering increased dissipation.
| [
{
"version": "v1",
"created": "Mon, 30 Dec 2024 17:43:57 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Berx",
"Jonas",
""
],
[
"Singh",
"Prashant",
""
],
[
"Proesmans",
"Karel",
""
]
] |
2501.00235 | Tongseok Lim | Daeyoung Jeong, Tongseok Lim and Euncheol Shin | Robust Intervention in Networks | The latest version of the paper can be accessed at
https://tlim0213.github.io/folder/papers/Robust_Intervention_in_Networks.pdf,
and the online appendix is available at
https://tlim0213.github.io/folder/papers/online_appendix.pdf | null | null | null | econ.TH | http://creativecommons.org/licenses/by/4.0/ | In economic settings such as learning, social behavior, and financial
contagion, agents interact through interdependent networks. This paper examines
how a decision maker (DM) can design an optimal intervention strategy under
network uncertainty, modeled as a zero-sum game against an adversarial
``Nature'' that reconfigures the network within an uncertainty set. Using
duality, we characterize the DM's unique robust intervention and identify the
worst-case network structure, which exhibits a rank-1 property, concentrating
risk along the intervention strategy. We analyze the costs of robustness,
distinguishing between global and local uncertainty, and examine the role of
higher-order uncertainties in shaping intervention outcomes. Our findings
highlight key trade-offs between maximizing influence and mitigating
uncertainty, offering insights into robust decision-making. This framework has
applications in policy design, economic regulation, and strategic interventions
in dynamic networks, ensuring their resilience against uncertainty in network
structures.
| [
{
"version": "v1",
"created": "Tue, 31 Dec 2024 03:00:40 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 05:24:50 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Jeong",
"Daeyoung",
""
],
[
"Lim",
"Tongseok",
""
],
[
"Shin",
"Euncheol",
""
]
] |
2501.01291 | Yu-Han Huang | Yu-Han Huang, Argyrios Gerogiannis, Subhonmesh Bose, Venugopal V.
Veeravalli | Detection Augmented Bandit Procedures for Piecewise Stationary MABs: A
Modular Approach | 34 pages, 2 figures, 1 table, submitted to JMLR | null | null | null | cs.AI cs.LG cs.SY eess.SY stat.ML | http://creativecommons.org/licenses/by/4.0/ | Conventional Multi-Armed Bandit (MAB) algorithms are designed for stationary
environments, where the reward distributions associated with the arms do not
change with time. In many applications, however, the environment is more
accurately modeled as being nonstationary. In this work, piecewise stationary
MAB (PS-MAB) environments are investigated, in which the reward distributions
associated with a subset of the arms change at some change-points and remain
stationary between change-points. Our focus is on the asymptotic analysis of
PS-MABs, for which practical algorithms based on change detection (CD) have
been previously proposed. Our goal is to modularize the design and analysis of
such CD-based Bandit (CDB) procedures. To this end, we identify the
requirements for stationary bandit algorithms and change detectors in a CDB
procedure that are needed for the modularization. We assume that the rewards
are sub-Gaussian. Under this assumption and a condition on the separation of
the change-points, we show that the analysis of CDB procedures can indeed be
modularized, so that regret bounds can be obtained in a unified manner for
various combinations of change detectors and bandit algorithms. Through this
analysis, we develop new modular CDB procedures that are order-optimal. We
compare the performance of our modular CDB procedures with various other
methods in simulations.
| [
{
"version": "v1",
"created": "Thu, 2 Jan 2025 15:18:18 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 19:35:40 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Huang",
"Yu-Han",
""
],
[
"Gerogiannis",
"Argyrios",
""
],
[
"Bose",
"Subhonmesh",
""
],
[
"Veeravalli",
"Venugopal V.",
""
]
] |
2501.01441 | Aditya Bhattacharya | Aditya Bhattacharya, Simone Stumpf, Robin De Croon, Katrien Verbert | Explanatory Debiasing: Involving Domain Experts in the Data Generation
Process to Mitigate Representation Bias in AI Systems | Pre-print version, please cite the main article instead of the
pre-print version | ACM CHI 2025 | 10.1145/3706598.3713497 | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | Representation bias is one of the most common types of biases in artificial
intelligence (AI) systems, causing AI models to perform poorly on
underrepresented data segments. Although AI practitioners use various methods
to reduce representation bias, their effectiveness is often constrained by
insufficient domain knowledge in the debiasing process. To address this gap,
this paper introduces a set of generic design guidelines for effectively
involving domain experts in representation debiasing. We instantiated our
proposed guidelines in a healthcare-focused application and evaluated them
through a comprehensive mixed-methods user study with 35 healthcare experts.
Our findings show that involving domain experts can reduce representation bias
without compromising model accuracy. Based on our findings, we also offer
recommendations for developers to build robust debiasing systems guided by our
generic design guidelines, ensuring more effective inclusion of domain experts
in the debiasing process.
| [
{
"version": "v1",
"created": "Thu, 26 Dec 2024 14:14:48 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 08:45:05 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Bhattacharya",
"Aditya",
""
],
[
"Stumpf",
"Simone",
""
],
[
"De Croon",
"Robin",
""
],
[
"Verbert",
"Katrien",
""
]
] |
2501.01821 | Aobo Kong | Aobo Kong, Wentao Ma, Shiwan Zhao, Yongbin Li, Yuchuan Wu, Ke Wang,
Xiaoqian Liu, Qicheng Li, Yong Qin, Fei Huang | SDPO: Segment-Level Direct Preference Optimization for Social Agents | null | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Social agents powered by large language models (LLMs) can simulate human
social behaviors but fall short in handling complex social dialogues. Direct
Preference Optimization (DPO) has proven effective in aligning LLM behavior
with human preferences across various agent tasks. However, standard DPO
focuses solely on individual turns, which limits its effectiveness in
multi-turn social interactions. Several DPO-based multi-turn alignment methods
with session-level data have shown potential in addressing this problem.While
these methods consider multiple turns across entire sessions, they are often
overly coarse-grained, introducing training noise, and lack robust theoretical
support. To resolve these limitations, we propose Segment-Level Direct
Preference Optimization (SDPO), which dynamically select key segments within
interactions to optimize multi-turn agent behavior. SDPO minimizes training
noise and is grounded in a rigorous theoretical framework. Evaluations on the
SOTOPIA benchmark demonstrate that SDPO-tuned agents consistently outperform
both existing DPO-based methods and proprietary LLMs like GPT-4o, underscoring
SDPO's potential to advance the social intelligence of LLM-based agents. We
release our code and data at
https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/SDPO.
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 14:09:46 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 05:42:17 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Kong",
"Aobo",
""
],
[
"Ma",
"Wentao",
""
],
[
"Zhao",
"Shiwan",
""
],
[
"Li",
"Yongbin",
""
],
[
"Wu",
"Yuchuan",
""
],
[
"Wang",
"Ke",
""
],
[
"Liu",
"Xiaoqian",
""
],
[
"Li",
"Qicheng",
""
],
[
"Qin",
"Yong",
""
],
[
"Huang",
"Fei",
""
]
] |
2501.01824 | Jingbo Sun | Jingbo Sun, Hengxiao Guo, Minfeng Gu, Ya-Ping Li, Yongjun Chen, D.
Gonz\'alez-Buitrago, Jian-Guo Wang, Sha-Sha Li, Hai-Cheng Feng, Dingrong
Xiong, Yanan Wang, Qi Yuan, Jun-Jie Jin, Wenda Zhang, Hongping Deng, and
Minghao Zhang | AT2021aeuk: A Repeating Partial Tidal Disruption Event Candidate in a
Narrow-line Seyfert 1 galaxy | null | null | null | null | astro-ph.HE astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A black hole (BH) can tear apart a star that ventures within its tidal
radius, producing a luminous flare as the stellar debris falls back, known as a
tidal disruption event (TDE). While TDEs in quiescent galaxies are relatively
well understood, identifying TDEs in active galactic nuclei (AGN) still remains
a significant challenge. We present the discovery of AT2021aeuk, a transient
exhibiting dual flares within around three years in a narrow-line Seyfert 1
galaxy. Multi-wavelength observations triggered during the second flare in 2023
revealed an extraordinary X-ray V-shaped light curve, strongly anti-correlated
with the optical light curve and accompanied by a lag of $\sim$40 days. This
behavior is inconsistent with both supernova and pure AGN origins. In addition,
a new broad component emerges in the Balmer lines during the second flare,
showing a clear reverberation signal to the continuum variation. We propose
that the dual-flare may be linked to a repeating partial TDE (rpTDE), where the
second flare results from a collision between the TDE stream and the inner
accretion disk, triggering an optical flare while simultaneously partially
destroying the X-ray corona. However, other mechanisms, such as a stellar-mass
BH (sBH) merger within an accretion disk, could produce similar phenomena,
which we cannot entirely rule out. The Vera C. Rubin Observatory will be a
powerful tool for further investigating the nature of such events in the
future.
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 14:15:25 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 05:56:37 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Sun",
"Jingbo",
""
],
[
"Guo",
"Hengxiao",
""
],
[
"Gu",
"Minfeng",
""
],
[
"Li",
"Ya-Ping",
""
],
[
"Chen",
"Yongjun",
""
],
[
"González-Buitrago",
"D.",
""
],
[
"Wang",
"Jian-Guo",
""
],
[
"Li",
"Sha-Sha",
""
],
[
"Feng",
"Hai-Cheng",
""
],
[
"Xiong",
"Dingrong",
""
],
[
"Wang",
"Yanan",
""
],
[
"Yuan",
"Qi",
""
],
[
"Jin",
"Jun-Jie",
""
],
[
"Zhang",
"Wenda",
""
],
[
"Deng",
"Hongping",
""
],
[
"Zhang",
"Minghao",
""
]
] |
2501.02053 | Shahar Hod | Shahar Hod | A sufficient condition for the development of superradiant instabilities
in charged black-hole spacetimes | 9 pages. arXiv admin note: text overlap with arXiv:2401.07907 | Journal of High Energy Physics 02, 197 (2025) | null | null | gr-qc astro-ph.HE hep-th | http://creativecommons.org/licenses/by/4.0/ | The physical and mathematical properties of charged black holes that are
linearly coupled to charged massive scalar fields are studied analytically. In
particular, we prove that, in the eikonal large-mass regime $M\mu\gg1$, the
compact dimensionless inequality $\Phi_{\text{H}}>Q/M$ provides a sufficient
condition for the development of superradiant instabilities in the curved
black-hole spacetime [here $\{M,Q,\Phi_{\text{H}}\}$ are respectively the mass,
the electric charge, and the horizon electrostatic potential of the central
black hole and $\mu$ is the proper mass of the field]. The familiar charged
Reissner-Nordstr\"om black hole does not satisfy this inequality. On the other
hand, we explicitly prove that all charged Ay\'on-Beato-Garc\'ia (ABG)
black-hole spacetimes satisfy this analytically derived sufficient condition
and may therefore become superradiantly unstable to perturbations of charged
massive scalar fields.
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 19:00:01 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Hod",
"Shahar",
""
]
] |
2501.02454 | Shunzhuang Huang | Shunzhuang Huang, Xinran Li and Panos Toulis | Finite-Sample Valid Randomization Tests for Monotone Spillover Effects | null | null | null | null | stat.ME stat.AP | http://creativecommons.org/licenses/by-sa/4.0/ | Randomization tests have gained popularity for causal inference under network
interference because they are finite-sample valid with minimal assumptions.
However, existing procedures are limited as they primarily focus on the
existence of spillovers through sharp null hypotheses on potential outcomes. In
this paper, we expand the scope of randomization procedures in network settings
by developing new tests for the monotonicity of spillover effects. These tests
offer insights into whether spillover effects increase, decrease, or exhibit
``diminishing returns'' along certain network dimensions of interest. Our
approach partitions the network into multiple (possibly overlapping) parts and
tests a monotone contrast hypothesis in each sub-network. The test decisions
can then be aggregated in various ways depending on how each test is
constructed. We demonstrate our method by re-analyzing a large-scale policing
experiment in Colombia, which reveals evidence of monotonicity related to the
``crime displacement hypothesis''. Our analysis suggests that crime spillovers
on a control street increase with the number of nearby streets receiving more
intense policing but diminish at higher exposure levels.
| [
{
"version": "v1",
"created": "Sun, 5 Jan 2025 06:30:42 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 05:30:43 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Huang",
"Shunzhuang",
""
],
[
"Li",
"Xinran",
""
],
[
"Toulis",
"Panos",
""
]
] |
2501.02792 | Liudong Chen | Liudong Chen, Jay Sethuraman, Bolun Xu | Gaming on Coincident Peak Shaving: Equilibrium and Strategic Behavior | null | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Power system operators and electric utility companies often charge consumers
a peak demand charge when the aggregate system demand reaches its maximum, a
practice known as coincident peak shaving. These charges incentivize consumers
to reduce usage during critical periods, alleviating stress on electricity
transmission and distribution systems, while also helping to recover the grid
investment costs. In this paper, we analyze the problem through the lens of
game theory, developing a model that captures how strategic consumer behavior
influences overall system efficiency. Our results reveal that the coincident
peak shaving game can be concave, quasiconcave with discontinuities, or
non-concave with discontinuities, depending on the extent of consumers'
demand-shifting capabilities. In a two-agent, two-period framework, we derive
closed-form Nash equilibrium solutions for each scenario and extend our
analysis to multi-agent settings. We prove the stability of these equilibrium
points and propose an algorithm to compute equilibrium outcomes across all game
configurations. Furthermore, we show that while the decentralized game model
achieves peak-shaving performance comparable to a centralized approach, it does
so at the cost of increased anarchy. In scenarios characterized by
quasi-concave and non-concave conditions, our analytical results demonstrate
that anarchy grows with greater consumer flexibility and disparities in
marginal shifting costs, and we explore how the number of agents affects system
efficiency. Finally, numerical simulations are provided to validate our
theoretical findings.
| [
{
"version": "v1",
"created": "Mon, 6 Jan 2025 06:25:46 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Jan 2025 17:58:19 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Jan 2025 03:32:35 GMT"
},
{
"version": "v4",
"created": "Thu, 27 Feb 2025 18:23:42 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Chen",
"Liudong",
""
],
[
"Sethuraman",
"Jay",
""
],
[
"Xu",
"Bolun",
""
]
] |
2501.03241 | Lea Gassab | Lea Gassab, Onur Pusuluk, Marco Cattaneo and \"Ozg\"ur E.
M\"ustecapl{\i}o\u{g}lu | Quantum Models of Consciousness from a Quantum Information Science
Perspective | null | Entropy 2025, 27, 243 | 10.3390/e27030243 | null | q-bio.NC quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This perspective explores various quantum models of consciousness from the
viewpoint of quantum information science, offering potential ideas and
insights. The models under consideration can be categorized into three distinct
groups based on the level at which quantum mechanics might operate within the
brain: those suggesting that consciousness arises from electron delocalization
within microtubules inside neurons, those proposing it emerges from the
electromagnetic field surrounding the entire neural network, and those positing
it originates from the interactions between individual neurons governed by
neurotransmitter molecules. Our focus is particularly on the Posner model of
cognition, for which we provide preliminary calculations on the preservation of
entanglement of phosphate molecules within the geometric structure of Posner
clusters. These findings provide valuable insights into how quantum information
theory can enhance our understanding of brain functions.
| [
{
"version": "v1",
"created": "Fri, 20 Dec 2024 06:31:03 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 20:56:08 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Gassab",
"Lea",
""
],
[
"Pusuluk",
"Onur",
""
],
[
"Cattaneo",
"Marco",
""
],
[
"Müstecaplıoğlu",
"Özgür E.",
""
]
] |
2501.03982 | Nick Koning | Nick W. Koning and Sam van Meer | Free Anytime Validity by Sequentializing a Test and Optional
Continuation with Tests as Future Significance Levels | Changes: added one-sided sequentialised t-test plus general strategy
for monotone likelihood ratios, added connection to tracking strong p-value,
added switching significance levels, more explicitly discourage randomization
and clarified relationship to randomization, generally improved writing | null | null | null | math.ST stat.ME stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anytime valid sequential tests permit us to stop and continue testing based
on the current data, without invalidating the inference. Given a maximum number
of observations $N$, one may believe this must come at the cost of power when
compared to a conventional test that waits until all $N$ observations have
arrived. Our first contribution is to show that this is false: for any valid
test based on $N$ observations, we derive an anytime valid sequential test that
matches it after $N$ observations. Our second contribution is that the outcome
of a continuously-interpreted test can be used as a significance level in
subsequent testing, leading to an overall procedure that is valid at the
original significance level. Combined this shows both the value of
continuously-interpreted tests, and the fact that anytime validity and optional
continuation are readily available in traditional testing, without requiring
explicit use of e-values. We illustrate this by deriving the anytime valid
sequentialized $z$-test and $t$-test, which at time $N$ coincide with the
traditional $z$-test and $t$-test. Lastly, we show the popular log-optimal
sequential $z$-test can be interpreted as desiring a rejection by the
traditional $z$-test at some tiny significance level in the distant future.
| [
{
"version": "v1",
"created": "Tue, 7 Jan 2025 18:39:28 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Jan 2025 15:27:28 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 13:50:49 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Koning",
"Nick W.",
""
],
[
"van Meer",
"Sam",
""
]
] |
2501.05955 | Leonid Polterovich | Michael Entov, Leonid Polterovich, Lenya Ryzhik | A Contact Topological Glossary for Non-Equilibrium Thermodynamics | 24 pages, 5 figures; revised and considerably expanded | null | null | null | math.SG math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We discuss the occurrence of some notions and results from contact topology
in the non-equilibrium thermodynamics. This includes the Reeb chords and the
partial order on the space of Legendrian submanifolds.
| [
{
"version": "v1",
"created": "Fri, 10 Jan 2025 13:33:41 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 23:45:53 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Entov",
"Michael",
""
],
[
"Polterovich",
"Leonid",
""
],
[
"Ryzhik",
"Lenya",
""
]
] |
2501.06022 | Denitsa Staicova | Denitsa Staicova | Modern Bayesian Sampling Methods for Cosmological Inference: A
Comparative Study | 16 pages, Final published version, | Universe 2025, 11(2), 68; | 10.3390/universe11020068 | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a comprehensive comparison of different Markov Chain Monte Carlo
(MCMC) sampling methods, evaluating their performance on both standard test
problems and cosmological parameter estimation. Our analysis includes
traditional Metropolis-Hastings MCMC, Hamiltonian Monte Carlo (HMC), slice
sampling, nested sampling as implemented in dynesty, and PolyChord. We examine
samplers through multiple metrics including runtime, memory usage, effective
sample size, and parameter accuracy, testing their scaling with dimension and
response to different probability distributions. While all samplers perform
well with simple Gaussian distributions, we find that HMC and nested sampling
show advantages for more complex distributions typical of cosmological
problems. Traditional MCMC and slice sampling become less efficient in higher
dimensions, while nested methods maintain accuracy but at higher computational
cost. In cosmological applications using BAO data, we observe similar patterns,
with particular challenges arising from parameter degeneracies and poorly
constrained parameters.
| [
{
"version": "v1",
"created": "Fri, 10 Jan 2025 15:01:20 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Feb 2025 09:05:44 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Staicova",
"Denitsa",
""
]
] |
2501.06139 | Vipin Sudevan | Reyhan D. Lambaga, Vipin Sudevan and Pisin Chen | SkyReconNet: A Cross-Resolution Contextual Integration Framework for
Inpainting with Application to Enhanced CMB Map Reconstruction | null | null | null | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a novel neural network, SkyReconNet, which combines the expanded
receptive fields of dilated convolutional layers along with standard
convolutions, to capture both the global and local features for reconstructing
the missing information in an image. We implement our network to inpaint the
masked regions in a full-sky Cosmic Microwave Background (CMB) map. Inpainting
CMB maps is a particularly formidable challenge when dealing with extensive and
irregular masks, such as galactic masks which can obscure substantial fractions
of the sky. The hybrid design of SkyReconNet leverages the strengths of
standard and dilated convolutions to accurately predict CMB fluctuations in the
masked regions, by effectively utilizing the information from surrounding
unmasked areas. During training, the network optimizes its weights by
minimizing a composite loss function that combines the Structural Similarity
Index Measure (SSIM) and mean squared error (MSE). SSIM preserves the essential
structural features of the CMB, ensuring an accurate and coherent
reconstruction of the missing CMB fluctuations, while MSE minimizes the
pixel-wise deviations, enhancing the overall accuracy of the predictions. The
predicted CMB maps and their corresponding angular power spectra align closely
with the targets, achieving the performance limited only by the fundamental
uncertainty of cosmic variance. The network's generic architecture enables
application to other physics-based challenges involving data with missing or
defective pixels, systematic artefacts etc. Our results demonstrate its
effectiveness in addressing the challenges posed by large irregular masks,
offering a significant inpainting tool not only for CMB analyses but also for
image-based experiments across disciplines where such data imperfections are
prevalent.
| [
{
"version": "v1",
"created": "Fri, 10 Jan 2025 18:00:30 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 16:59:14 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Lambaga",
"Reyhan D.",
""
],
[
"Sudevan",
"Vipin",
""
],
[
"Chen",
"Pisin",
""
]
] |
2501.06188 | Ramisa Rahman | Ramisa Akther Rahman, Joshua Bennett Lovell, Eric W. Koch, David J.
Wilner, Sean M. Andrews, Kristina Monsch, Dan Ha | The Radiowave Hunt for Young Stellar Object Emission and Demographics
(RADIOHEAD): A Radio Luminosity${-}$Spectral Type Dependence in
Taurus${-}$Auriga YSOs | Accepted for publication in ApJ. 26 pages, inc. 12 figures, 3 tables | null | 10.3847/1538-4357/ad9f37 | null | astro-ph.SR astro-ph.EP astro-ph.GA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We measure the radio continuum fluxes at the locations of all
Gaia${-}$confirmed members of Taurus${-}$Auriga using Karl G. Jansky Very Large
Array Sky Survey data (VLASS; 2${-}$4 GHz, $\sigma_{\rm{VLASS}}{\sim}110{-}140
\mu$Jy, $2.5''$ resolution) spanning 3 VLASS epochs (2019, 2021, and 2023). We
present 35 detections coincident with young Taurus${-}$Auriga stars (29 in
individual VLASS images, 6 via stacking). We find a strong dependence on
spectral type, wherein the fractional detection rate of radio emission
coincident with early-type young stellar objects (YSOs) is systematically
higher than late-type YSOs, ranging from 25% $\pm$ 13% for B${-}$F YSOs, 21%
$\pm$ 11% for G YSOs, 18.4% $\pm$ 6.3% for K0${-}$K4 YSOs, 15.5% $\pm$ 5.4% for
K5${-}$K9 YSOs, 7.0% $\pm$ 2.7% for M0${-}$M2 YSOs, 2.3% $\pm$ 0.9% for
M3${-}$M6 YSOs, and 1.9% $\pm$ 1.9% for YSOs with SpTs later than M7. We
present cumulative density distributions of radio luminosity densities that
demonstrate a significant luminosity enhancement for early- versus late-type
YSOs. We find 25% of the detected sources to be significantly variable. We
discuss possible interpretations of this dependence, which may reflect stellar
magnetic activity, binary interactions, or stellar flaring. We find that
mid-infrared YSO class is a strong indicator of radio detectability consistent
with higher frequency Taurus-Auriga VLA surveys, with class III stars detected
at a rate of 8.8% $\pm$ 1.6%, class IIs at 2.0% $\pm$ 1.2%, and combined class
0s, Is and Fs at 8.0% $\pm$ 5.4%.
| [
{
"version": "v1",
"created": "Fri, 10 Jan 2025 18:59:58 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Rahman",
"Ramisa Akther",
""
],
[
"Lovell",
"Joshua Bennett",
""
],
[
"Koch",
"Eric W.",
""
],
[
"Wilner",
"David J.",
""
],
[
"Andrews",
"Sean M.",
""
],
[
"Monsch",
"Kristina",
""
],
[
"Ha",
"Dan",
""
]
] |
2501.06195 | Riccardo Droghei PhD | Riccardo Droghei | Deformed Boson Algebras and $\mathcal{W}_{\alpha,\beta,\nu}$-Coherent
States: A New Quantum Framework | 18 pages, 6 figures | Mathematics, 13(5), 759 (2025) | 10.3390/math13050759 | null | math.QA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce a novel class of coherent states, termed
$\mathcal{W}^{(\bar{\alpha},\bar{\nu})}(z)$-coherent states, constructed using
a deformed boson algebra based on the generalized factorial
$[n]_{\alpha,\beta,\nu}!$. This algebra extends conventional factorials,
incorporating advanced special functions such as the Mittag-Leffler and Wright
functions, enabling the exploration of a broader class of quantum states. The
mathematical properties of these states, including their continuity,
completeness, and quantum fluctuations, are analyzed. A key aspect of this work
is the resolution of the Stieltjes moment problem associated with these states,
achieved through the inverse Mellin transformation method. The framework
provides insights into the interplay between classical and quantum regimes,
with potential applications in quantum optics and fractional quantum mechanics.
By extending the theoretical landscape of coherent states, this study opens
avenues for further exploration in mathematical physics and quantum
technologies.
| [
{
"version": "v1",
"created": "Thu, 26 Dec 2024 00:00:16 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 14:46:07 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Droghei",
"Riccardo",
""
]
] |
2501.06733 | Renrui Qi | Renrui Qi | Notes on Laver Tables | Thanks hyp_cos for finding some typos in the previous version | null | null | null | math.LO math.CO | http://creativecommons.org/licenses/by/4.0/ | We present some new lower bound estimates for certain numbers in Laver table
theory and introduce several related structures of interest.
| [
{
"version": "v1",
"created": "Sun, 12 Jan 2025 07:08:03 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Feb 2025 04:33:07 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 13:48:49 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Qi",
"Renrui",
""
]
] |
2501.06738 | Sharif Ahmed | Sharif Ahmed, Nasir U. Eisty | Hold On! Is My Feedback Useful? Evaluating the Usefulness of Code Review
Comments | Accepted for publication in Empirical Software Engineering (EMSE)
Journal | null | 10.1007/s10664-025-10617-1 | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Context: In collaborative software development, the peer code review process
proves beneficial only when the reviewers provide useful comments. Objective:
This paper investigates the usefulness of Code Review Comments (CR comments)
through textual feature-based and featureless approaches. Method: We select
three available datasets from both open-source and commercial projects.
Additionally, we introduce new features from software and non-software domains.
Moreover, we experiment with the presence of jargon, voice, and codes in CR
comments and classify the usefulness of CR comments through featurization,
bag-of-words, and transfer learning techniques. Results: Our models outperform
the baseline by achieving state-of-the-art performance. Furthermore, the result
demonstrates that the commercial gigantic LLM, GPT-4o, or non-commercial naive
featureless approach, Bag-of-Word with TF-IDF, is more effective for predicting
the usefulness of CR comments. Conclusion: The significant improvement in
predicting usefulness solely from CR comments escalates research on this task.
Our analyses portray the similarities and differences of domains, projects,
datasets, models, and features for predicting the usefulness of CR comments.
| [
{
"version": "v1",
"created": "Sun, 12 Jan 2025 07:22:13 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Ahmed",
"Sharif",
""
],
[
"Eisty",
"Nasir U.",
""
]
] |
2501.06791 | Pushpendra Singh | Dilpreet Kaur, Pushpendra Singh | Classification of simple quandles of small order | v1 extended to simple quandles | null | null | null | math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this article, we define quasiprimitive quandles and describe them with the
help of quasiprimitive permutation groups. As a consequence, we enumerate
finite non-affine simple quandles up to order $4096$.
| [
{
"version": "v1",
"created": "Sun, 12 Jan 2025 12:05:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 14:58:27 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Kaur",
"Dilpreet",
""
],
[
"Singh",
"Pushpendra",
""
]
] |
2501.06838 | Du Chen | Du Chen, Liyi Chen, Zhengqiang Zhang, Lei Zhang | Generalized and Efficient 2D Gaussian Splatting for Arbitrary-scale
Super-Resolution | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Implicit Neural Representation (INR) has been successfully employed for
Arbitrary-scale Super-Resolution (ASR). However, INR-based models need to query
the multi-layer perceptron module numerous times and render a pixel in each
query, resulting in insufficient representation capability and computational
efficiency. Recently, Gaussian Splatting (GS) has shown its advantages over INR
in both visual quality and rendering speed in 3D tasks, which motivates us to
explore whether GS can be employed for the ASR task. However, directly applying
GS to ASR is exceptionally challenging because the original GS is an
optimization-based method through overfitting each single scene, while in ASR
we aim to learn a single model that can generalize to different images and
scaling factors. We overcome these challenges by developing two novel
techniques. Firstly, to generalize GS for ASR, we elaborately design an
architecture to predict the corresponding image-conditioned Gaussians of the
input low-resolution image in a feed-forward manner. Each Gaussian can fit the
shape and direction of an area of complex textures, showing powerful
representation capability. Secondly, we implement an efficient differentiable
2D GPU/CUDA-based scale-aware rasterization to render super-resolved images by
sampling discrete RGB values from the predicted continuous Gaussians. Via
end-to-end training, our optimized network, namely GSASR, can perform ASR for
any image and unseen scaling factors. Extensive experiments validate the
effectiveness of our proposed method. The code and models will be released.
| [
{
"version": "v1",
"created": "Sun, 12 Jan 2025 15:14:58 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Jan 2025 14:09:23 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 09:07:43 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Chen",
"Du",
""
],
[
"Chen",
"Liyi",
""
],
[
"Zhang",
"Zhengqiang",
""
],
[
"Zhang",
"Lei",
""
]
] |
2501.07375 | Tongyu Wu | Tongyu Wu, Changhao Miao, Yuntian Zhang, Fang Deng and Chen Chen | A RankNet-Inspired Surrogate-Assisted Hybrid Metaheuristic for Expensive
Coverage Optimization | null | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Coverage optimization generally involves deploying a set of facilities to
best satisfy the demands of specified points, with broad applications in fields
such as location science and sensor networks. Recent applications reveal that
the subset site selection coupled with continuous angular parameter
optimization can be formulated as Mixed-Variable Optimization Problems (MVOPs).
Meanwhile, high-fidelity discretization and visibility analysis significantly
increase computational cost and complexity, evolving the MVOP into an Expensive
Mixed-Variable Optimization Problem (EMVOP). While canonical Evolutionary
Algorithms have yielded promising results, their reliance on numerous fitness
evaluations is too costly for our problem. Furthermore, most surrogate-assisted
methods face limitations due to their reliance on regression-based models. To
address these issues, we propose the RankNet-Inspired Surrogate-assisted Hybrid
Metaheuristic (RI-SHM), an extension of our previous work. RI-SHM integrates
three key components: (1) a RankNet-based pairwise global surrogate that
innovatively predicts rankings between pairs of individuals, bypassing the
challenges of fitness estimation in discontinuous solution space; (2) a
surrogate-assisted local Estimation of Distribution Algorithm that enhances
local exploitation and helps escape from local optima; and (3) a fitness
diversity-driven switching strategy that dynamically balances exploration and
exploitation. Experiments demonstrate that our algorithm can effectively handle
large-scale coverage optimization tasks of up to 300 dimensions and more than
1,800 targets within desirable runtime. Compared to state-of-the-art algorithms
for EMVOPs, RI-SHM consistently outperforms them by up to 56.5$\%$ across all
tested instances.
| [
{
"version": "v1",
"created": "Mon, 13 Jan 2025 14:49:05 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 04:32:30 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Wu",
"Tongyu",
""
],
[
"Miao",
"Changhao",
""
],
[
"Zhang",
"Yuntian",
""
],
[
"Deng",
"Fang",
""
],
[
"Chen",
"Chen",
""
]
] |
2501.07665 | Manuel \'Ubeda-Flores | Ana P\'erez, Mercedes Prieto-Alaiz, Fernando Chamizo, Eckhard
Liebscher, Manuel \'Ubeda-Flores | Nonparametric estimation of the multivariate Spearman's footrule: a
further discussion | A thorough review of the work is required to correct errors and
improve clarity | Fuzzy Sets and Systems, volume 467 (2023), article 108489 | 10.1016/j.fss.2023.02.010 | null | math.ST math.PR stat.TH | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this paper, we propose two new estimators of the multivariate rank
correlation coefficient Spearman's footrule which are based on two general
estimators for Average Orthant Dependence measures. We compare the new
proposals with a previous estimator existing in the literature and show that
the three estimators are asymptotically equivalent, but, in small samples, one
of the proposed estimators outperforms the others. We also analyse Pitman
efficiency of these indices to test for multivariate independence as compared
to multivariate versions of Kendall's tau and Spearman's rho.
| [
{
"version": "v1",
"created": "Mon, 13 Jan 2025 19:52:32 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 10:06:47 GMT"
}
] | 2025-02-28T00:00:00 | [
[
"Pérez",
"Ana",
""
],
[
"Prieto-Alaiz",
"Mercedes",
""
],
[
"Chamizo",
"Fernando",
""
],
[
"Liebscher",
"Eckhard",
""
],
[
"Úbeda-Flores",
"Manuel",
""
]
] |
Subsets and Splits