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

Feb 5

Submitted by

taesiri

ERNIE 5.0 Technical Report

·
438 authors

Submitted by

xiaochonglinghu

FASA: Frequency-aware Sparse Attention

alibaba-inc

Submitted by

zelaix

WideSeek-R1: Exploring Width Scaling for Broad Information Seeking via Multi-Agent Reinforcement Learning

RLinf

Submitted by

shrango

Training Data Efficiency in Multimodal Process Reward Models

·
7 authors

Submitted by

Jungang

OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models

·
15 authors

1

Submitted by

LongMountain

HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing

XiaomiMiMo

Submitted by

tellarin

EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models

·
6 authors

1

Submitted by

xihc-ucb

Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

UCBerkeley

University of California, Berkeley

Submitted by

beatccjiang

TIDE: Trajectory-based Diagnostic Evaluation of Test-Time Improvement in LLM Agents

·
10 authors

Submitted by

UML

SoMA: A Real-to-Sim Neural Simulator for Robotic Soft-body Manipulation

Submitted by

yuezhouhu

Residual Context Diffusion Language Models

UCBerkeley

University of California, Berkeley

Submitted by

QPHutu

Rethinking the Trust Region in LLM Reinforcement Learning

sail

1

Submitted by

zooblastlbz

Semantic Routing: Exploring Multi-Layer LLM Feature Weighting for Diffusion Transformers

KlingTeam

Submitted by

taesiri

HY3D-Bench: Generation of 3D Assets

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

WENGSYX

AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations

WestlakeNLP

Text Intelligence Lab of Westlake University

Submitted by

baohao

Self-Hinting Language Models Enhance Reinforcement Learning

MicrosoftResearch

Microsoft Research

Submitted by

CheeryLJH

Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestration

NJU-LINK

Submitted by

Ablustrund

CL-bench: A Benchmark for Context Learning

tencent

Submitted by

Ayanami0730

A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces

muset-ai

Submitted by

Jiafei1224

VLS: Steering Pretrained Robot Policies via Vision-Language Models

allenai

Ai2

Submitted by

jmhb

PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR

·
8 authors

Submitted by

sijunhe

MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering

ernie-research

Submitted by

Ningyu

From Data to Behavior: Predicting Unintended Model Behaviors Before Training

Zhejiang University

Submitted by

yasNing

Agent-Omit: Training Efficient LLM Agents for Adaptive Thought and Observation Omission via Agentic Reinforcement Learning

Didichuxing

Submitted by

bowiehsu

D-CORE: Incentivizing Task Decomposition in Large Reasoning Models for Complex Tool Use

alibaba-inc

Submitted by

Tyrannosaurus

Horizon-LM: A RAM-Centric Architecture for LLM Training

notredame

University of Notre Dame

Submitted by

fdu-lcz

BatCoder: Self-Supervised Bidirectional Code-Documentation Learning via Back-Translation

·
12 authors

1

Submitted by

taesiri

Likelihood-Based Reward Designs for General LLM Reasoning

metaresearch

Submitted by

azminetoushikwasi

SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?

ciol-research

Computational Intelligence and Operations Laboratory (CIOL)

Submitted by

Zijie-chen

A2Eval: Agentic and Automated Evaluation for Embodied Brain

X-Humanoid

Submitted by

shuaishuaicdp

Quantifying the Gap between Understanding and Generation within Unified Multimodal Models

umd-zhou-lab

Submitted by

271754echo

Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition

·
8 authors

Submitted by

dimakarp1996

No One-Size-Fits-All: Building Systems For Translation to Bashkir, Kazakh, Kyrgyz, Tatar and Chuvash Using Synthetic And Original Data

·
1 authors

1

Submitted by

HangGuo

Efficient Autoregressive Video Diffusion with Dummy Head

MicrosoftResearch

Microsoft Research

Submitted by

taesiri

Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging

·
5 authors

Submitted by

hansenhua

Context Learning for Multi-Agent Discussion

Tsinghua University

Submitted by

weblzw

Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models

·
5 authors

Submitted by

taesiri

Protein Autoregressive Modeling via Multiscale Structure Generation

ByteDance-Seed

Submitted by

DipanAI

SAFE: Stable Alignment Finetuning with Entropy-Aware Predictive Control for RLHF

·
1 authors

Submitted by

thebajajra

RexBERT: Context Specialized Bidirectional Encoders for E-commerce

owlgebra-ai

1

Submitted by

linzheng

Proxy Compression for Language Modeling

·
5 authors

Submitted by

RuijieZhu

SkeletonGaussian: Editable 4D Generation through Gaussian Skeletonization

ustc

university of science and technology of china

Submitted by

Jinnkunn

"I May Not Have Articulated Myself Clearly": Diagnosing Dynamic Instability in LLM Reasoning at Inference Time

·
4 authors

Submitted by

PeterLauLukCh

Reward-free Alignment for Conflicting Objectives

·
4 authors

Submitted by

hzp

LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization

·
9 authors

1

Submitted by

abdevd

FOTBCD: A Large-Scale Building Change Detection Benchmark from French Orthophotos and Topographic Data

retgenai

Submitted by

Snarcy

OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysis

·
3 authors