Failure Detection on Close Box Out-of-Distribution (train test)
100TPRTemporal Non-Distr. Min.
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Temporal Non-Distr. Min.Category=Diffusion2026.03 | 100 | 27 | 12.35 | |
| DDPM LossCategory=Diffusion2026.03 | 100 | 94 | 14.54 | |
| Temporal DDPM LossCategory=Diffusion2026.03 | 100 | 37 | 13.66 | |
| Policy EncoderCategory=Embed.2026.03 | 100 | 0 | 1.59 | |
| CLIP PretrainedCategory=Embed.2026.03 | 100 | 0 | 8.2 | |
| ResNet PretrainedCategory=Embed.2026.03 | 100 | 0 | 15.51 | |
| GPT-4o Image QACategory=VLM2026.03 | 100 | 0 | 23.2 | |
| Sentinel (STAC MMD* + GPT-4o Video QA*)Configuration=Parallel Ensemble2026.03 | 100 | 90 | 14.25 | |
| STAC For. KLCategory=STAC2026.03 | 99 | 85 | 14.04 | |
| STAC MMD*Category=STAC2026.03 | 99 | 93 | 14.72 | |
| STAC Rev. KLCategory=STAC2026.03 | 93 | 97 | 15.12 | |
| Temporal Diffusion Recon.Category=Diffusion2026.03 | 92 | 100 | 15.75 | |
| GPT-4o Video QA*Category=VLM2026.03 | 69 | 95 | 21.02 | |
| Diffusion Recon.Category=Diffusion2026.03 | 40 | 100 | 17.08 | |
| Diffusion Output VarianceCategory=Diffusion2026.03 | 28 | 100 | 17.27 |