Skip to content

project

MAPoRL

Method for training multiple language-model agents together with reinforcement learning so they improve at collaborative discussion, published at ACL 2025.

Current clusters