| title | AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research |
|---|---|
| authors | Marjan Moodi, Xuankang Zhu, Fernando De Mesentier Silva, Harold Chaput, Mohammad Reza Taesiri |
| date | 2026-07-20 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2608.11216 |
| upvotes | 14 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | World modeling is an unsettled field: architectures, training objectives, and state representations interact in complex ways, and no single recipe dominates across environments. This makes it an ideal testbed for AI coding agents acting as autonomous researchers--a setting in which the improvement direction is not specified in advance, unlike the engineering-to-spec tasks that dominate current agent benchmarks. We introduce AutoWorldModel-Bench, a closed-loop benchmark in which frontier coding agents autonomously improve a provided world-model starter under a fixed compute budget. The benchmark spans eight game environments under a unified structured-state representation--ground-truth entity state extracted from each game and consumed through a shared tensor format--which isolates dynamics modeling from perception and enables minutes-per-run iteration. Across 64 sessions, Codex-5.4 and Claude Opus 4.6 improve their starter on 63; in 91% of sessions the winning edit is a non-trivial research-style modification--a new objective, representation, rollout procedure, or architectural change--rather than a hyperparameter tweak. Our benchmark offers a setting in which frontier coding agents can be evaluated on open-ended research rather than engineering-to-spec problems. |
cite
@misc{moodi2026autoworldmodel,
title={AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research},
author={Marjan Moodi and Xuankang Zhu and Fernando De Mesentier Silva and Harold Chaput and Mohammad Reza Taesiri},
year={2026},
eprint={2608.11216},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2608.11216}
}