16 papers
Latent On-Policy Self-Distillation
Guibin Zhang, Jiayang Lyu, Ran Sun +4
Enabling agents to learn from experience and internalize it into their policy has become a central problem in self-evolving AI. On-policy self-distillation (OPSD) offers an effecti…
SafeSieve: From Heuristics to Experience in Progressive Pruning for LLM-based Multi-Agent Communication
Ruijia Zhang, Xinyan Zhao, Ruixiang Wang +5
LLM-based multi-agent systems exhibit strong collaborative capabilities but often suffer from redundant communication and excessive token overhead. Existing methods typically enhan…
AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents
Hanjun Luo, Shenyu Dai, Chiming Ni +5
Despite the rapid advancement of LLM-based agents, the reliable evaluation of their safety and security remains a significant challenge. Existing rule-based or LLM-based evaluators…
Visual Multi-Agent System: Mitigating Hallucination Snowballing via Visual Flow
Xinlei Yu, Chengming Xu, Guibin Zhang +8
Multi-Agent System (MAS) powered by Visual Language Models (VLMs) enables challenging tasks but suffers from a novel failure term, multi-agent visual hallucination snowballing, whe…
MemGen: Weaving Generative Latent Memory for Self-Evolving Agents
Guibin Zhang, Muxin Fu, Shuicheng Yan
Agent memory shapes how Large Language Model (LLM)-powered agents, akin to the human brain, progressively refine themselves through environment interactions. Existing paradigms rem…
OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment
Liang Lin, Zhihao Xu, Junhao Dong +10
Large language model (LLM) alignment faces a critical dilemma when addressing multiple human preferences: improvements in one dimension frequently come at the expense of others, cr…