Lessons Learned: A Multi-Agent Framework for Code LLMs to Learn and Improve

Jie Chen (Hong Kong Baptist University) · Yuanzhe Liu (Rensselaer Polytechnic Institute) · Ryan Deng (Massachusetts Institute of Technology) · Tim Kaler (Massachusetts Institute of Technology) · Xuhao Chen (Massachusetts Institute of Technology) · Charles Leiserson (Massachusetts Institute of Technology) · Yao Ma (Rensselaer Polytechnic Institute)

agent collaborationagent performance improvementbanking selection mechanismcode optimizationcollective solution processcomplementary strengths knowledge transferlesson solicitationlesson-based frameworkmulti-agent learningmulti-llm collaborationoptimization categoriesperformance granularitysmall llmstask specialization

Recent studies show that LLMs possess different skills and specialize in different tasks. In fact, we observe that their varied performance occur in several levels of granularity. For example, in the code optimization task, code LLMs excel at different optimization categories and no one dominates others. This observation prompts the question of how one leverages multiple LLM agents to solve a coding problem without knowing their complementary strengths a priori. We argue that a team of agents can learn from each other's successes and failures so as to improve their own performance. Thus, a lesson is the knowledge produced by an agent and passed on to other agents in the collective solution process. We propose a lesson-based collaboration framework, design the lesson solicitation--banking--selection mechanism, and demonstrate that a team of small LLMs with lessons learned can outperform a much larger LLM and other multi-LLM collaboration methods.