# Failure Detection on Close Box Combined Source: [Failure Detection on Close Box Combined]() Collection index: [llms.txt]() Data updated: 2026-03-13T07:48:16.309Z Evaluation of Failure Detection performance using the Close Box Combined dataset, reporting metrics such as True Positive Rate (TPR), True Negative Rate (TNR), and Accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Failure Detection | | Dataset (including any stated split) | Close Box Combined | | Metric | TPR | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [TPR](): Higher is better; unit: PERCENT. - [TNR](): Higher is better; unit: PERCENT. - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | TPR | Result date | Report paper | | --- | --- | --- | --- | | Temporal Non-Distr. Min. | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | DDPM Loss (Eq. A.1) | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | Temporal DDPM Loss | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | CLIP Pretrained | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | ResNet Pretrained | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | GPT-4o Image QA | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | Gemini 1.5 Pro Image QA | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | Gemini 1.5 Pro Video QA | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | Sentinel | 100 | Not recorded | 69b399e80636e5bf4afbc8cd | | STAC For. KL | 99 | Not recorded | 69b399e80636e5bf4afbc8cd | | STAC MMD\* | 99 | Not recorded | 69b399e80636e5bf4afbc8cd | | Policy Encoder | 94 | Not recorded | 69b399e80636e5bf4afbc8cd | | STAC Rev. KL | 93 | Not recorded | 69b399e80636e5bf4afbc8cd | | Temporal Diffusion Recon. | 92 | Not recorded | 69b399e80636e5bf4afbc8cd | | GPT-4o Video QA\* | 77 | Not recorded | 69b399e80636e5bf4afbc8cd | | Claude 3.5 Sonnet Image QA | 77 | Not recorded | 69b399e80636e5bf4afbc8cd | | Claude 3.5 Sonnet Video QA | 55 | Not recorded | 69b399e80636e5bf4afbc8cd | | Diffusion Recon. | 37 | Not recorded | 69b399e80636e5bf4afbc8cd | | Diffusion Output Variance | 26 | Not recorded | 69b399e80636e5bf4afbc8cd | ### Temporal Non-Distr. Min. TPR: 100. Result ID: 69b3c07a47f34ce9249d7d2d. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. ### DDPM Loss (Eq. A.1) TPR: 100. Result ID: 69b3c07a47f34ce9249d7d30. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. ### Temporal DDPM Loss TPR: 100. Result ID: 69b3c07a47f34ce9249d7d31. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. ### CLIP Pretrained TPR: 100. Result ID: 69b3c07a47f34ce9249d7d34. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Failure Detector Category":"Embed."}. This record does not assert independent reproduction. ### ResNet Pretrained TPR: 100. Result ID: 69b3c07a47f34ce9249d7d35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"Embed."}. This record does not assert independent reproduction. ### GPT-4o Image QA TPR: 100. Result ID: 69b3c07a47f34ce9249d7d39. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Gemini 1.5 Pro Image QA TPR: 100. Result ID: 69b3c07a47f34ce9249d7d3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Gemini 1.5 Pro Video QA TPR: 100. Result ID: 69b3c07a47f34ce9249d7d3c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Sentinel TPR: 100. Result ID: 69b3c07a47f34ce9249d7d3f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"Combined","Configuration":"STAC MMD\* + GPT-4o Video QA\*"}. This record does not assert independent reproduction. ### STAC For. KL TPR: 99. Result ID: 69b3c07a47f34ce9249d7d36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"STAC"}. This record does not assert independent reproduction. ### STAC MMD\* TPR: 99. Result ID: 69b3c07a47f34ce9249d7d38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"STAC"}. This record does not assert independent reproduction. ### Policy Encoder TPR: 94. Result ID: 69b3c07a47f34ce9249d7d33. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Failure Detector Category":"Embed."}. This record does not assert independent reproduction. ### STAC Rev. KL TPR: 93. Result ID: 69b3c07a47f34ce9249d7d37. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"STAC"}. This record does not assert independent reproduction. ### Temporal Diffusion Recon. TPR: 92. Result ID: 69b3c07a47f34ce9249d7d2f. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. ### GPT-4o Video QA\* TPR: 77. Result ID: 69b3c07a47f34ce9249d7d3a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Claude 3.5 Sonnet Image QA TPR: 77. Result ID: 69b3c07a47f34ce9249d7d3d. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Claude 3.5 Sonnet Video QA TPR: 55. Result ID: 69b3c07a47f34ce9249d7d3e. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"VLM"}. This record does not assert independent reproduction. ### Diffusion Recon. TPR: 37. Result ID: 69b3c07a47f34ce9249d7d2e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. ### Diffusion Output Variance TPR: 26. Result ID: 69b3c07a47f34ce9249d7d32. [Reporting paper]() Recorded evaluation configuration: {"Failure Detector Category":"Diffusion"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.