jocelynd/playworld-bench

PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives Project | Paper | Code | Leaderboard Video world models generate future states from an initial observation and user actions. Comparing interactive models fairly is difficult because the actions required to reach the same long-horizon objective can vary substantially across models. PlayWorld evaluates world models from the perspective of a human player. Each case provides an initial world, a… See the full description on the dataset page: https://huggingface.co/datasets/jocelynd/playworld-bench.

<iframe src="https://silytics.si/embed/item/d97078df-003c-4aa5-a505-c8ab863eb1b8" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.