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

Aug 26

Submitted by

lyhisme

Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs

·
7 authors

Submitted by

JUNJIE99

WeMM-Embedding: WeChat Multi-Modal Embedding Technical Report

tencent

Submitted by

pshlego

AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Traces

microsoft

Submitted by

WeiChow

On-Policy Self-Distillation in Diffusion Models

ByteDance-Seed

Submitted by

Lingaaaaaaa

Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses

princetonu

Princeton University

Submitted by

taesiri

On-policy Distillation with Verifiable Reward

·
8 authors

Submitted by

HaojunChen

From Seeing to Acting: Smart Glasses as First-Person Intelligence Platforms

·
3 authors

Submitted by

QPHutu

Best Practice Critic Optimization

NUS-DSA3101

National University of Singapore

Submitted by

taesiri

LAION-BVD: A 10-Million-Hour Open Video Dataset for Multimodal Pre-training

·
12 authors

Submitted by

taesiri

Meta^n: Recursive Self-Improvement through Emergent Depth

·
4 authors

Submitted by

taesiri

CAFE: Self-Improving Search Agents Need Co-Evolving Feedback

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

YanZhanPKU

Length-Adaptive Decoding for Masked Diffusion Machine Translation

PekingUniversity

Peking University

Submitted by

kaizhao

TorchMorph: CUDA-accelerated Morphological Transforms

intcomp

Intelligent Computing Laboratory