List your company

Failure Detection on Close Box In-Distribution

100TPRTemporal Non-Distr. Min.
2242.2562.582.75Mar 12, 2026

Evaluation results

MethodLinks
Temporal Non-Distr. Min.Failure Detector Category=Diffusion2026.03100975
Temporal Diffusion Recon.Failure Detector Category=Diffusion2026.03100968.47
DDPM Loss (Eq. A.1)Failure Detector Category=Diffusion2026.03100908.27
Temporal DDPM LossFailure Detector Category=Diffusion2026.03100957.53
CLIP PretrainedFailure Detector Category=Embed.2026.031009515.73
ResNet PretrainedFailure Detector Category=Embed.2026.031009517.87
STAC For. KLFailure Detector Category=STAC2026.03100906.6
STAC Rev. KLFailure Detector Category=STAC2026.03100957.6
STAC MMD*Failure Detector Category=STAC2026.03100947.2
GPT-4o Image QAFailure Detector Category=VLM2026.03100023.2
GPT-4o Video QA*Failure Detector Category=VLM2026.031008921.2
Gemini 1.5 Pro Image QAFailure Detector Category=VLM2026.03100021.2
Gemini 1.5 Pro Video QAFailure Detector Category=VLM2026.031005717.2
Claude 3.5 Sonnet Image QAFailure Detector Category=VLM2026.03100623.2
SentinelFailure Detector Category=Combined, Configuration=STAC MMD* + GPT-4o Video QA*2026.03100865.47
Claude 3.5 Sonnet Video QAFailure Detector Category=VLM2026.03833123.2
Diffusion Recon.Failure Detector Category=Diffusion2026.03339513.6
Diffusion Output VarianceFailure Detector Category=Diffusion2026.03339414
Policy EncoderFailure Detector Category=Embed.2026.03259816.27