ziyjiang/WorldAuditBench

WorldAuditBench Interactive 3D World Auditing with Multimodal Agents 213 tasks · 13 environments · 5 anomaly familiesUC Santa Barbara · MIT CSAIL · MIT–IBM Watson AI Lab WorldAuditBench evaluates whether multimodal agents can explore a 3D world, investigate suspicious observations, and identify anomalies with visual evidence. An agent may need to approach an object, test a collision, change its viewpoint, or revisit a location to establish what is wrong. Representative… See the full description on the dataset page: https://huggingface.co/datasets/ziyjiang/WorldAuditBench.

<iframe src="https://silytics.si/embed/item/83ebe841-a4c0-40a3-a438-ffc78f57533a" 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 Information may be incomplete or out of date.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.