new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

byAK and the research community

Oct 24

Submitted by

Mqleet

Human-Agent Collaborative Paper-to-Page Crafting for Under $0.1

Submitted by

yuezhouhu

AdaSPEC: Selective Knowledge Distillation for Efficient Speculative Decoders

GeorgiaTech

Georgia Institute of Technology

Submitted by

taesiri

Open-o3 Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence

ByteDance

Submitted by

taesiri

HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives

antgroup

Submitted by

GuyYariv

DyPE: Dynamic Position Extrapolation for Ultra High Resolution Diffusion

The Hebrew University of Jerusalem

Submitted by

jojo0217

Loopholing Discrete Diffusion: Deterministic Bypass of the Sampling Wall

·
5 authors

2

Submitted by

zenyn

SAKE: Towards Editing Auditory Attribute Knowledge of Large Audio-Language Models

nationaltaiwan

Submitted by

yudian

Every Question Has Its Own Value: Reinforcement Learning with Explicit Human Values

tencent

Submitted by

umarbutler

The Massive Legal Embedding Benchmark (MLEB)

isaacus

Submitted by

zenyn

Investigating Safety Vulnerabilities of Large Audio-Language Models Under Speaker Emotional Variations

nationaltaiwan

Submitted by

Necolizer

Search Self-play: Pushing the Frontier of Agent Capability without Supervision

Quark-LLM

Submitted by

zcxu-eric

Seed3D 1.0: From Images to High-Fidelity Simulation-Ready 3D Assets

ByteDance-Seed

Submitted by

yujiazheng

Thought Communication in Multiagent Collaboration

·
7 authors

2

Submitted by

RUBBISHLIKE

Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence

PekingUniversity

Peking University

Submitted by

taesiri

LayerComposer: Interactive Personalized T2I via Spatially-Aware Layered Canvas

snap-research

Submitted by

Xiaolong-Wang

ARGenSeg: Image Segmentation with Autoregressive Image Generation Model

inclusionAI

2

Submitted by

taesiri

AlphaFlow: Understanding and Improving MeanFlow Models

snap-research

Submitted by

jenyag

Diff-XYZ: A Benchmark for Evaluating Diff Understanding

JetBrains-Research

JetBrains Research

2

Submitted by

BryanW

From Masks to Worlds: A Hitchhiker's Guide to World Models

·
10 authors

Submitted by

NaiveUser

Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs

amazon

2

Submitted by

taesiri

ImpossibleBench: Measuring LLMs' Propensity of Exploiting Test Cases

·
3 authors

Submitted by

eaglew

CiteGuard: Faithful Citation Attribution for LLMs via Retrieval-Augmented Validation

·
5 authors

Submitted by

BUILDERlym

Communication to Completion: Modeling Collaborative Workflows with Intelligent Multi-Agent Communication

·
9 authors

Submitted by

ParagonLight

Adamas: Hadamard Sparse Attention for Efficient Long-Context Inference

·
7 authors

2

Submitted by

ParagonLight

Long-Context Attention Benchmark: From Kernel Efficiency to Distributed Context Parallelism

·
7 authors

2

Submitted by

aritraroy24

ComProScanner: A multi-agent based framework for composition-property structured data extraction from scientific literature

slimeslab

South London Innovative Materials Evaluation Squad (SLIMES) Lab

2

Submitted by

Snooow1029

MSC-Bench: A Rigorous Benchmark for Multi-Server Tool Orchestration

nationaltaiwan

2

Submitted by

ravfogs

Emergence of Linear Truth Encodings in Language Models

·
5 authors