AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Unsupervised Partner Design Enables Robust Ad-hoc Teamwork

Constantin Ruhdorfer, Matteo Bortoletto, Victor Oei, Anna Penzkofer, Andreas Bulling

ICML 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Collaborative multi-agent perception · 0 citations (OpenAlex)

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 adaptive partner generation with learnability selection (UPD) instead of E3T with fixed epsilonE3T · Overcooked-AI (5 layouts avg)average return: 78.8 -> 94.4Improved
Use UPD instead of E3T across swept epsilon valuesE3T · Level-Based Foraging, ten evaluation partnersaverage episodic returnImproved
Add persistent action bias mask to partner generation (ablation)UPD · Overcooked-AI (5 layouts avg)average return: 87.9 -> 94.4Improved
Add learnability-based partner selection (ablation)UPD · Overcooked-AI (5 layouts avg)average return: 90.8 -> 94.4Improved
Add learnability selection on Cramped Room layout (ablation)UPD · Overcooked-AI Cramped Roomaverage return: 110.8 -> 107.5No clear effect
Sample epsilon from U(0,1) with random partner generation only (no learnability) (ablation)E3T · Overcooked-AI (5 layouts avg)average return: 78.8 -> 90.8Improved
Use joint partner-environment curriculum (JUPD) instead of SFL levels with E3T partnersSFLE3T · Overcooked Generalisation Challenge 5x5average return: 44.0 -> 58.9Improved
Use self-play on random levels (CEC) instead of joint curriculumJUPD · Overcooked Generalisation Challenge 5x5average return: 58.9 -> 23.9Hurt
Use UPD instead of population-based MEP and FCPFCP and MEP populations of 48 · Overcooked-AI (5 layouts avg)average return: 75.3 -> 94.4Improved
Train ego with UPD instead of SP, MEP, E3T with human partnersHuman-AI Overcooked-AI · Human study, 12 participants, AA/CR/CCreturn and survey ratingsImproved

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