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

May 28

Submitted by

Mithas-01

ProRL: Effective Reinforcement Learning for Proactive Recommendation via Rectified Policy Gradient Estimation

Fudan-University

Fudan University

Submitted by

taesiri

Agent Explorative Policy Optimization for Multimodal Agentic Reasoning

nvidia

Submitted by

Paranioar

From Pixels to Words -- Towards Native One-Vision Models at Scale

·
21 authors

Submitted by

taesiri

Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players

nvidia

Submitted by

Xkev

Self-Improving Language Models with Bidirectional Evolutionary Search

Harvard

Harvard University

Submitted by

amphora

ResearchMath-14K: Scaling Research-Level Mathematics via Agents

SeoulNatlUniv

Seoul National University

Submitted by

Ningyu

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

alibaba-inc

Submitted by

SII-Molu

DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes

Fudan-University

Fudan University

Submitted by

Zuyan

GEM: Generative Supervision Helps Embodied Intelligence

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

KangsanKim71

Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents

kaist-ai

Submitted by

memray

ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence

google

Submitted by

yunyangge

OSP-Next: Efficient High-Quality Video Generation with Sparse Sequence Parallelism, HiF8 Quantization, and Reinforcement Learning

PekingUniversity

Peking University

Submitted by

FlameF0X

Triplet-Block Diffusion RWKV

·
5 authors

Submitted by

Ningyu

Rethinking Memory as Continuously Evolving Connectivity

alibaba-inc

Submitted by

bowiehsu

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

alibaba-inc

Submitted by

wangxz098

Guiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders

·
7 authors

Submitted by

xiwenyoumu

Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving

nvidia

Submitted by

yasNing

HRBench: Benchmarking and Understanding Thinking-Mode Switch Strategies in Hybrid-Reasoning LLMs

tencent

Submitted by

CherryDurian

LiveBrowseComp: Are Search Agents Searching, or Just Verifying What They Already Know?

XiaohongshuAI

Submitted by

comin

OmniVerifier-M1: Multimodal Meta-Verifier with Explicit Structured Recalibration

·
10 authors

Submitted by

josephziheng

Less is More: Early Stopping Rollout for On-Policy Distillation

UCLA

University of California, Los Angeles

Submitted by

ryancll118

GE-Sim 2.0: A Roadmap Towards Comprehensive Closed-loop Video World Simulators for Robotic Manipulation

agibot-world

Submitted by

gonzmart

Lost in Sampling: Assessing Lexical Reachability in LLMs via the Word Coverage Score (WCS)

·
6 authors

Submitted by

yflantmy

SkillGrad: Optimizing Agent Skills Like Gradient Descent

·
5 authors

Submitted by

Luckyyy

VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild

rednote-hilab

Submitted by

MarkWang

Joint Training of Multi-Token Prediction in Reinforcement Learning via Optimal Coefficient Calibration

·
6 authors

Submitted by

taesiri

CubePart: An Open-Vocabulary Part-Controllable 3D Generator

·
12 authors

Submitted by

nicolepcx

AgensFlow: A Coordination-Policy Substrate for Multi-Agent Systems

·
1 authors

Submitted by

amphora

Chartographer: Counterfactual Chart Generation for Evaluating Vision-Language Models

·
4 authors

Submitted by

amazingj

ESC-Skills: Discovering and Self-Evolving Skills for Emotional Support Conversations

DianJin

Submitted by

YangyiH

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective

SphereLab

Submitted by

yixuantt

AI Research Agents Narrow Scientific Exploration

·
2 authors

Submitted by

anmolagarwal999

Verus-SpecGym: An Agentic Environment for Evaluating Specification Autoformalization

CarnegieMellonU

Carnegie Mellon University

Submitted by

chengq9

Advancing Creative Physical Intelligence in Large Multimodal Models

UIUC-CS

University of Illinois at Urbana-Champaign

Submitted by

namespace-ERI

AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning

·
8 authors

Submitted by

taesiri

AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation

Harvard

Harvard University

Submitted by

phuongnm

Revealing Algorithmic Deductive Circuits for Logical Reasoning

·
3 authors

Submitted by

voice-biomarkers

Clark Hash: Stateless Sparse Johnson-Lindenstrauss Quantization for Neural Embeddings

·
2 authors

Submitted by

taesiri

PEAM: Parametric Embodied Agent Memory through Contrastive Internalization of Experience in Minecraft

·
5 authors