Zhenfei Yin

Eigen-Agent: Adaptive Multi-Agent Scientific Reasoning with Monitor-Based RAG

Interleaving Reasoning for Better Text-to-Image Generation

LatentEvolve: Self-Evolving Test-Time Scaling in Latent Space

CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards

DIVA: Discrete Diffusion Vision-Language-Action Models for Parallelized Action Generation

VeriWeb: Verifiable Long-Chain Web Benchmark for Agentic Information-Seeking

Better know nothing than half-know anything: A Precise and Efficient Dataset for Scientific Reasoning in Language Models

A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory

CompBench: Benchmarking Complex Instruction-guided Image Editing

MASLab: A Unified and Comprehensive Codebase for LLM-based Multi-Agent Systems

Ego to World: Collaborative Spatial Reasoning in Embodied Systems via Reinforcement Learning

Diversity for The Win: Towards Building Multi-Agent Systems with Heterogeneous LLMs

VFXMaster: Unlocking Dynamic Visual Effect Generation via In-Context Learning

RoboMonster: Compositional Generalization of Heterogeneous Multi-End Effector Embodied Agents

Diagnose, Localize, Align: A Full-Stack Framework for Reliable LLM Multi-Agent Systems under Instruction Conflicts

Scaling Behaviors of LLM Reinforcement Learning Post-Training: An Empirical Study in Mathematical Reasoning

Building Scalable Real-World Robot Data Generation via Compositional Simulation

LiveSearchBench: An Automatically Constructed Benchmark for Retrieval and Reasoning over Dynamic Knowledge

Actial: Activate Spatial Reasoning Ability of Multimodal Large Language Models

WorldSimBench: Towards Video Generation Models as World Simulators

WorldSimBench: Towards Video Generation Models as World Simulators

SPA-VL: A Comprehensive Safety Preference Alignment Dataset for Vision Language Model

MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems

OASIS: Open Agents Social Interaction Simulations on a Large Scale

Two Heads Are Better Than One: A Multi-Agent System Has the Potential to Improve Scientific Idea Generation