#artificial intelligence
Sep 2026
Improving LLM Collaboration via Multi-Agent Preference Learning
This work instantiates MAPL using MARL from human feedback with a learned reward model and multi-agent direct preference optimization (MADPO), and shows that MAPL can improve collaboration quality and efficiency while approaching the performance of MARL with fixed, well-defined rewards.
Shuo Liu, Xinzichen Li, Tian-Le Chen et al.
· arXiv.org · 0 citations