中文

LLM-ALSO: LLM-Driven Adaptive Learning-Signal Optimization for Multi-Agent Reinforcement Learning

多智能体系统 2026-05-29 v1

摘要

Effective training-time guidance is central to multi-agent reinforcement learning (MARL), yet remains difficult in sparse-reward settings where weak supervision limits coordination and policy improvement, and existing methods often require substantial domain expertise or manual design effort. Large language models (LLMs) provide a promising alternative for flexible learning-signal design, yet existing LLM-based methods remain largely single-agent-oriented, one-shot, or weakly validated for the evolving training dynamics of cooperative MARL. To address these limitations, we propose LLM-ALSO, an iterative LLM-driven adaptive learning-signal optimization framework for MARL. Rather than directly deploying LLM-generated rewards, LLM-ALSO decomposes adaptation into iterative diagnosis, proposal, and validation: a Critic LLM diagnoses stage-specific learning and coordination failures from sparse-return metrics and compact behavior evidence, a Generator LLM proposes candidate reward-shaping configurations conditioned on the diagnosis, and branch-validation feedback refines candidates before they affect the main training trajectory. Through short-horizon validation and stage-aware adaptation, LLM-ALSO promotes only validated updates into training, reducing the risk of unreliable LLM-generated modifications. Experiments on sparse-reward cooperative MARL tasks show that LLM-ALSO improves sparse-evaluation performance and learning efficiency.

关键词

引用

@article{arxiv.2605.29293,
  title  = {LLM-ALSO: LLM-Driven Adaptive Learning-Signal Optimization for Multi-Agent Reinforcement Learning},
  author = {Xiaoguang Wu and Zhi Zheng and Hui Xiong},
  journal= {arXiv preprint arXiv:2605.29293},
  year   = {2026}
}

备注

14 pages, 6 figures, 6 tables