AI Paper Atlas

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

Benchmarking World-Model Learning with Environment-Level Queries

Archana Warrier, Dat Nguyen, Michelangelo Naim, Moksh Jain, Yichao Liang, Karen Schroeder, Cambridge Yang, Joshua B. Tenenbaum, Sebastian Josef Vollmer, Kevin Ellis, Zenna Tavares

ICML 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: World-model-based agents

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
humans versus reasoning models on the same AutumnBench tasksReasoning models (Claude 4 Sonnet, Gemini 2.5 Pro/Flash, o3, Qwen3) · AutumnBench (129 tasks)score: Humans outperformed all five reasoning models across every environment and task typeHurt
stochastic instead of deterministic environmentsReasoning models · AutumnBench (17 of 43 stochastic)score: Reasoning models did significantly better in stochastic settings than in deterministic onesImproved
stochastic instead of deterministic environmentsHumans · AutumnBench (17 of 43 stochastic)score: Humans performed nearly identically in both deterministic and stochastic environmentsNo clear effect
spend more compute per problem (higher-cost model)Reasoning models · AutumnBench masked-frame prediction and planningscore: higher cost improved performance in 37% of masked-frame prediction and planning tasksMixed
spend more compute per problem (higher-cost model)Reasoning models · AutumnBench change detectionscore: 33% for change detectionMixed
spend more compute per problem across environmentsReasoning models · AutumnBench (43 environments)score: 25 environments improved monotonically with cost; 18 environments gave no benefit from additional computeMixed
use reset as an exploration action (human-style) instead of rarelyHumans vs reasoning models · AutumnBenchreset fraction of actions: Humans 12.5% resets and no-ops each; every model under 7% combined, Claude 2.1%Improved
systematic replay of action subsequences around resets (human behaviour)Humans vs reasoning models · AutumnBenchLCS ratio pre/post reset: human mean 0.637 vs. 0.188--0.360 for modelsImproved

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