让多个大模型讨论时保持思路一致，避免互相干扰。
Context Learning for Multi-Agent Discussion
- 为每个模型训练上下文生成器，动态调整讨论内容。
- 在复杂任务上性能提升20%至50%，更易达成共识。
- 适合需要多智能体协作推理的场景，如学术解题与机器人控制。
多智能体讨论（MAD）近期受到广泛关注，多个大语言模型通过结构化讨论协同解决问题。然而，我们发现当前方法常因各模型间上下文不一致，导致讨论失序、无法达成连贯结论。本文提出多大模型上下文学习方法（M2CL），为每个代理训练上下文生成器，能基于自动信息组织与优化，在每轮讨论中动态生成上下文指令。受理论洞察启发，M2CL通过自适应机制训练生成器，以控制上下文一致性与输出差异。该机制使模型避免过早收敛于多数噪声，逐步达成正确共识。我们在学术推理、具身任务和移动端控制等挑战性任务上评估，结果表明M2CL性能显著优于现有方法20%–50%，同时具备良好可迁移性与计算效率。
原文摘要 · Abstract (English)
Multi-Agent Discussion (MAD) has garnered increasing attention very recently, where multiple LLM instances collaboratively solve problems via structured discussion. However, we find that current MAD methods easily suffer from discussion inconsistency, LLMs fail to reach a coherent solution, due to the misalignment between their individual contexts.In this paper, we introduce a multi-LLM context learning method (M2CL) that learns a context generator for each agent, capable of dynamically generating context instructions per discussion round via automatic information organization and refinement. Specifically, inspired by our theoretical insights on the context instruction, M2CL train the generators to control context coherence and output discrepancies via a carefully crafted self-adaptive mechanism.It enables LLMs to avoid premature convergence on majority noise and progressively reach the correct consensus. We evaluate M2CL on challenging tasks, including academic reasoning, embodied tasks, and mobile control. The results show that the performance of M2CL significantly surpasses existing methods by 20%--50%, while enjoying favorable transferability and computational efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。