benchmarks evals
3D-World Auditors Found at Most 42.3 Percent
WorldAuditBench made agents navigate, inspect and validate 213 anomalies across thirteen simulated environments.
Summary
WorldAuditBench made agents navigate, inspect and validate 213 anomalies across thirteen simulated environments.
WorldAuditBench couples two tasks: moving through a simulated 3D world to gather evidence and reasoning over what the agent sees to identify anomalies such as floating objects or traversable walls. Five frontier models were evaluated under a fixed exploration budget in two agent designs. Reported success ranged from 6.6% to 42.3%, compared with 83.4% for people. The gap shows that interactive evidence gathering remains much harder than recognizing a defect in a supplied frame.
Why it matters
WorldAuditBench made agents navigate, inspect and validate 213 anomalies across thirteen simulated environments.
Limits and context
No additional limitation was separately recorded.
Key claims
WorldAuditBench made agents navigate, inspect and validate 213 anomalies across thirteen simulated environments.
Evidence: source-2026-10-01-012
Sources
- arXiv preprint 2609.40325arXiv · primary research
Corrections
No corrections have been recorded for this story.