List your company

Task Progression Failure Detection on Cover Object Policy Success Rate: 3% (Out-of-Distribution)

100TPRGPT-4o Image QA
-4235077Mar 12, 2026

Evaluation results

MethodLinks
GPT-4o Image QAmodality=Image QA2026.03100011.03
GPT-4o Video QAmodality=Video QA2026.03100010.06
Gemini 1.5 Pro Video QAmodality=Video QA2026.0310007.54
Claude 3.5 Sonnet Video QA + Success Videomodality=Video QA, context=Success Video2026.039410011.59
Gemini 1.5 Pro Image QAmodality=Image QA2026.0391011.15
Sentinel (STAC MMD* + Claude 3.5 Sonnet Prompt Ensemble*)components=STAC MMD* + Claude 3.5 Sonnet Prompt Ensemble*2026.03881008.69
Claude 3.5 Sonnet Prompt Ensemble*strategy=Prompt Ensemble2026.038510012
Sentinel (STAC MMD* + Claude 3.5 Sonnet Video QA)components=STAC MMD* + Claude 3.5 Sonnet Video QA2026.03821008.68
Claude 3.5 Sonnet Video QAmodality=Video QA2026.037910012
Claude 3.5 Sonnet Video QA + Goal Imagesmodality=Video QA, context=Goal Images2026.037610011.74
Claude 3.5 Sonnet Image QAmodality=Image QA2026.037010012
STAC MMD*variant=MMD2026.0391007.73
Temporal Non-Distr. Min.2026.0361007.6
STAC For. KLvariant=Forward KL2026.0331006.4
STAC Rev. KLvariant=Reverse KL2026.0331006.4
Diffusion Output Variance2026.030100—