Siheng Chen

RATE-DISTORTION OPTIMIZED PRAGMATIC COMMUNICATION FOR COLLABORATIVE PERCEPTION

AgentFold: Long-Horizon Web Agents with Proactive Context Folding

Decentralized and Lifelong-Adaptive Multi-Agent Collaborative Learning

ChatAni: Language-Driven Multi-Actor Animation Generation in Street Scenes

MemGUI-Bench: Benchmarking Memory of Mobile GUI Agents in Dynamic Environments

MobileA3gent: Training Mobile GUI Agents Using Decentralized Self-Sourced Data from Diverse Users

InfoMosaic-Bench: Evaluating Multi-Source Information Seeking in Tool-Augmented Agents

EmboMatrix: A Scalable Training-Ground for Embodied Decision-Making

SWE-Dev: Training and Evaluating Autonomous End-to-End Feature-Driven Software Development

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

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

ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering

MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools

CodeAlign: Resolving Modality Isolation in Heterogeneous Collaborative Perception

Towards Scalable Web Browsing via Tool-Augmented Programmatic Agent Pair

SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents

Self-Evolving Multi-Agent Collaboration Networks for Software Development

Training-Free Message Passing for Learning on Hypergraphs

Point Cluster: A Compact Message Unit for Communication-Efficient Collaborative Perception

OOD-Barrier: Build a Middle-Barrier for Open-Set Single-Image Test Time Adaptation via Vision Language Models

Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models

BeliefMapNav: 3D Voxel-Based Belief Map for Zero-Shot Object Navigation

Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models

Synthesizing Post-Training Data for LLMs through Multi-Agent Simulation

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

Improving Defense Mechanisms for Subgraph-Structure Membership Inference Attacks

Federated Instruction Tuning of LLMs with Domain Coverage Augmentation