Search papers, labs, and topics across Lattice.
This paper introduces MASkills, a continual learning framework designed to enhance multi-agent LLM systems by optimizing agent skills, which serve as structured procedural knowledge for decision-making. The authors address the limitations of existing self-reflection methods by implementing a novel agent-optimization pipeline that incorporates skill-conditioned credit assignment, hierarchical credit aggregation, and momentum-smoothed optimization. Experimental results on tasks such as HotpotQA, LoCoMo, and GAIA show that MASkills significantly improves the performance and adaptability of multi-agent systems through effective skill refinement and management.
Multi-agent LLMs can now continually optimize their skills, leading to unprecedented adaptability and performance in complex tasks.
LLM-based multi-agent systems have shown strong performance on complex tasks, yet continual improvement from interaction experience remains challenging. Existing self-reflection methods build experience memories, but memories are mostly hard to invoke, refine, or scale, while agent skills offer a more actionable unit: structured procedural knowledge that specifies when to act, how to act, and which resources or tools to use. We introduce MASkills, a continual learning framework that optimizes multi-agent LLM systems through agent skills. MASkills presents a new agent-optimization pipeline that integrates skill-conditioned credit assignment, hierarchical credit aggregation, and momentum-smoothed optimization, enabling agent skill libraries to evolve through refinement, induction, consolidation, and pruning. Experiments on HotpotQA, LoCoMo, and GAIA demonstrate the effectiveness of MASkills across multiple agentic tasks. Our code is available at https://github.com/DaRL-GenAI/MASkills
