AI Paper Atlas

Research map / Multi-agent systems and games / Cooperative multi-agent RL

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

Zhuohui Zhang, Bin Cheng, Bin He

ICML 2026 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Offline and networked MARL

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GRPO with return-aware and executability reward after SFTDLM-SFT · SMAC corridortest win rate: 0.81 -> 0.92Improved
add GRPO after SFTDLM-SFT · SMAC 3s_vs_5ztest win rate: 0.84 -> 0.94Improved
add GRPO after SFTDLM-SFT · SMAC 5m_vs_6mtest win rate: 0.72 -> 0.80Improved
add GRPO after SFTDLM-SFT · SMAC zero-shot zerg_5_vs_5win rate: 0.42 -> 0.58Improved
add GRPO after SFTDLM-SFT · SMAC zero-shot 3mwin rate: 0.90 -> 0.93No clear effect
apply GRPO preference optimizationDLM-SFT · all SMAC tasksOOD action rateImproved
use dialogue-style inter-agent context with pre-trained LLM instead of MADTDLM-SFT vs MADT · SMAC zero-shotwin rate: 0.00 -> 0.59 (protoss_5_vs_5)Improved
use instruction-tuned LLM zero-shot (SayCan) instead of fine-tuningSayCan · SMACwin rate: 0.00 (corridor)Hurt
use LoRA instead of full fine-tuning in SFT stage (ablation)DLM-SFT · multi-task SMACrepresentation capacityHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy