DLM: Unified Decision Language Models for Offline Multi-Agent Sequential Decision Making

ArXi:2604.23557v1 Announce Type: cross Building scalable and reusable multi-agent decision policies from offline datasets remains a challenge in offline multi-agent reinforcement learning (MARL), as existing methods often rely on fixed observation formats and action spaces that limit generalization. In contrast, large language models (LLMs) offer a flexible modeling interface that can naturally accommodate heterogeneous observations and actions.