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

Jan 27

Submitted by

mhjiang0408

daVinci-Dev: Agent-native Mid-training for Software Engineering

GAIR

Submitted by

weizhoudb

Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs

SJTU

Shanghai Jiao Tong University

Submitted by

zptu

The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

ZetangForward

Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers

SUDA

Soochow University

Submitted by

LanguageBind

iFSQ: Improving FSQ for Image Generation with 1 Line of Code

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

LHL3341

Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility

SJTU

Shanghai Jiao Tong University

Submitted by

taesiri

Self-Refining Video Sampling

·
6 authors

Submitted by

Knykny

Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability

1

Submitted by

frontierai

VIBEVOICE-ASR Technical Report

microsoft

Submitted by

lgmeta

Paying Less Generalization Tax: A Cross-Domain Generalization Study of RL Training for LLM Agents

metaresearch

Submitted by

charlesdj

AR-Omni: A Unified Autoregressive Model for Any-to-Any Generation

ModalityDance

Submitted by

taesiri

SAGE: Steerable Agentic Data Generation for Deep Search with Execution Feedback

google

Submitted by

taesiri

Agentic Very Long Video Understanding

metaresearch

Submitted by

tiantiaf

End-to-End Joint ASR and Speaker Role Diarization with Child-Adult Interactions

·
5 authors

Submitted by

taesiri

SkyReels-V3 Technique Report

Skywork

Submitted by

SP2001

Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts

Salesforce

Submitted by

hankaixyz

Diffusion In Diffusion: Reclaiming Global Coherence in Semi-Autoregressive Diffusion

huawei-noah

HUAWEI Noah's Ark Lab

Submitted by

HongruCai

One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment

ModalityDance

Submitted by

HakHan

DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal

·
4 authors

Submitted by

jongwoopark7978

IVRA: Improving Visual-Token Relations for Robot Action Policy with Training-Free Hint-Based Guidance

·
6 authors

Submitted by

yumeow122

CGPT: Cluster-Guided Partial Tables with LLM-Generated Supervision for Table Retrieval

NchuNLPLab

NCHU Natural Language Processing Lab

Submitted by

smilestruggler

The Side Effects of Being Smart: Safety Risks in MLLMs' Multi-Image Reasoning

thu-coai

Conversational AI (CoAI) group from Tsinghua University

Submitted by

taesiri

UI Remix: Supporting UI Design Through Interactive Example Retrieval and Remixing

·
5 authors

Submitted by

Jize1

RouteMoA: Dynamic Routing without Pre-Inference Boosts Efficient Mixture-of-Agents

·
12 authors

Submitted by

rifqifarhansyah

PingPong: A Natural Benchmark for Multi-Turn Code-Switching Dialogues

·
9 authors

1