AI Paper Atlas

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

Role-Level Inductive Bias for Cross-Task Generalization in Multi-Agent Reinforcement Learning

Chang Yao, Youfang Lin, Shoucheng Song, Hao Wu, Shengkun Yang, Yuqing Ma, Kai Lv

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

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
remove UW-DPP diverse role selection (ablation)GTR · SMAC zero-shot 4s5z_vs_4s7ztest win rate: 0.08 -> 0.47 (adding UW-DPP)Improved
add intention prediction regularization (ablation)GTR · SMAC zero-shot 4s5z_vs_4s7ztest win rate: 0.09 -> 0.47Improved
add influence prediction regularization (ablation)GTR · SMAC zero-shot 4s5z_vs_4s7ztest win rate: 0.18 -> 0.47Improved
add role embedding in observation encoder (top-down perception) (ablation)GTR · SMAC zero-shot 4s5z_vs_4s7ztest win rate: 0.14 -> 0.47Improved
add role embedding in the mixer (ablation)GTR · SMAC zero-shot 4s5z_vs_4s7ztest win rate: 0.38 -> 0.47Improved
add role components on source tasks (ablation)GTR · SMAC source tasks (5m_vs_6m, 3s_vs_5z, 3s5z_vs_3s6z)test win rate: e.g. 5m_vs_6m 0.89 -> 0.92 (vs w/o intention)Improved
use GTR role-level bias instead of entity-level baselines (HPN, RPG, ASN, UPDeT)zero-shot transfer · SMAC unseen taskstest win rate: 8m vs 9m: 0.56 (RPG) -> 0.91; 4s vs 8z: 0.25 -> 0.72Improved
use GTR instead of baselines on source taskszero-shot transfer · SMAC source taskstest win rate: 5m vs 6m: 0.95 ± 0.08 (HPN) vs 0.92 ± 0.04No clear effect
initialize from pretrained GTR instead of training from scratchGTR · SMAC few-shot 4s5z_vs_4s7ztest win rateImproved

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