List your company

Task Progression Failure Detection on Close Box Out-of-Distribution

100TPRGPT-4o Image QA
25.1244.566483.44Mar 12, 2026

Evaluation results

MethodLinks
GPT-4o Image QAFailure Detector Category=VLM2026.03100—22.68
Gemini 1.5 Pro Video QAFailure Detector Category=VLM2026.0398—15.47
Gemini 1.5 Pro Image QAFailure Detector Category=VLM2026.0396—23.2
SentinelComponents=STAC MMD* + GPT-4o Video QA*2026.0396—12.2
GPT-4o Video QA*Failure Detector Category=VLM2026.0387—22
Claude 3.5 Sonnet Video QAFailure Detector Category=VLM2026.0380—23.2
Claude 3.5 Sonnet Image QAFailure Detector Category=VLM2026.0378—23.2
Temporal Non-Distr. Min.Failure Detector Category=Diff.2026.0367—7.46
STAC For. KLFailure Detector Category=STAC2026.0361—8.14
STAC Rev. KLFailure Detector Category=STAC2026.0361—10.11
STAC MMD*Failure Detector Category=STAC2026.0361—9.06
Diffusion Output VarianceFailure Detector Category=Diff.2026.0328—11.57