AI Paper Atlas

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

MangoBench: A Benchmark for Multi-Agent Goal-Conditioned Offline Reinforcement Learning

Yi Wang, Ningze Zhong, Zhiheng Fu, Longguang Wang, Ye Zhang, Yulan Guo

CVPR 2026 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms

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
use fully decentralized IHIQL instead of CTDE HIQL-CTDE (ablation)hierarchical goal-conditioned MARL · AntMaze-navigate large (2x4)success rate (%): 44.0 -> 85.4Improved
use fully decentralized IHIQL instead of CTDE HIQL-CTDE (ablation)hierarchical goal-conditioned MARL · AntMaze-navigate giant (2x4)success rate (%): 1.4 -> 57.3Improved
use CTDE instead of decentralized training for flat GCIVL variant (ablation)IGCIVL · AntMaze-navigate large (2x4)success rate (%): 15.0 -> 20.5No clear effect
use CTDE instead of decentralized training for flat GCIVL variant (ablation)IGCIVL · AntMaze-navigate medium (2x4)success rate (%): 76.0 -> 75.0No clear effect
evaluate with multiple goals instead of a single goalIHIQL · lift-barriersuccess rate: 78% -> 82%Improved
evaluate with multiple goals instead of a single goalGCMBC · lift-barriersuccess rate: 22% -> 47%Improved
evaluate with multiple goals instead of a single goalICRL · lift-barriersuccess rate: 37% -> 56%Improved
use shaped dense reward instead of sparse reward (ablation)OMIGA · antmaze-medium-navigatesuccess rateImproved
add goal-conditioning to offline MARL with sparse reward (GCOMIGA, GCOMAR)OMIGA and OMAR · MangoBench locomotion taskssuccess rateHurt
use vision observations instead of state observations (ablation)goal-conditioned MARL baselines · manipulation taskssuccess rateImproved
use multi-agent IHIQL instead of single-agent HIQLIHIQL · teleport mazessuccess rateImproved
use multi-agent baselines instead of single-agent for Ant-Soccermulti-agent baselines · Ant-Soccersuccess rateHurt

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