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

Jul 15

Submitted by

AaronHuangWei

EmbRACE-3K: Embodied Reasoning and Action in Complex Environments

·
9 authors

Submitted by

dorni

SpeakerVid-5M: A Large-Scale High-Quality Dataset for Audio-Visual Dyadic Interactive Human Generation

·
9 authors

Submitted by

Liang0223

LayerCake: Token-Aware Contrastive Decoding within Large Language Model Layers

·
10 authors

Submitted by

Word2Li

REST: Stress Testing Large Reasoning Models by Asking Multiple Problems at Once

·
8 authors

Submitted by

zsytony

CompassJudger-2: Towards Generalist Judge Model via Verifiable Rewards

·
5 authors

Submitted by

raymin0223

Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation

·
11 authors

Submitted by

chenguolin

MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second

·
7 authors

Submitted by

zsytony

Reasoning or Memorization? Unreliable Results of Reinforcement Learning Due to Data Contamination

·
12 authors

Submitted by

amphora

From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation

·
6 authors

Submitted by

Inoichan

A Practical Two-Stage Recipe for Mathematical LLMs: Maximizing Accuracy with SFT and Efficiency with Reinforcement Learning

·
3 authors