# Failure Detection on VLABench (Unseen Tasks) Source: [Failure Detection on VLABench (Unseen Tasks)]() Collection index: [llms.txt]() Data updated: 2026-06-01T08:18:13.516Z Evaluation of model performance on Failure Detection using the VLABench (Unseen Tasks) subset, measured by bACC, wACC, and TWA. 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) | VLABench (Unseen Tasks) | | Metric | bACC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [bACC](): Higher is better; unit: PERCENT. - [wACC](): Higher is better; unit: PERCENT. - [TWA](): Higher is better; unit: PERCENT. ## Recorded results | Method | bACC | Result date | Report paper | | --- | --- | --- | --- | | Hide-and-Seek | 71.3 | Not recorded | 6a1d11dbdbd7150506f83d90 | | Cosine k-NN | 70.4 | Not recorded | 6a1d11dbdbd7150506f83d90 | | LogpZO | 67.4 | Not recorded | 6a1d11dbdbd7150506f83d90 | | SAFE-MLP | 64.1 | Not recorded | 6a1d11dbdbd7150506f83d90 | | PCA KMeans | 61.6 | Not recorded | 6a1d11dbdbd7150506f83d90 | | Mahalanobis | 61.1 | Not recorded | 6a1d11dbdbd7150506f83d90 | | STAC | 59.1 | Not recorded | 6a1d11dbdbd7150506f83d90 | | ACE | 59.1 | Not recorded | 6a1d11dbdbd7150506f83d90 | | EigenScore | 56.6 | Not recorded | 6a1d11dbdbd7150506f83d90 | | RND | 55.1 | Not recorded | 6a1d11dbdbd7150506f83d90 | | Cluster Entropy | 52.6 | Not recorded | 6a1d11dbdbd7150506f83d90 | | SAFE-LSTM | 48.4 | Not recorded | 6a1d11dbdbd7150506f83d90 | ### Hide-and-Seek bACC: 71.3. Result ID: 6a1d4034fbc373592c165d97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### Cosine k-NN bACC: 70.4. Result ID: 6a1d4034fbc373592c165d8d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### LogpZO bACC: 67.4. Result ID: 6a1d4034fbc373592c165d90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### SAFE-MLP bACC: 64.1. Result ID: 6a1d4034fbc373592c165d96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### PCA KMeans bACC: 61.6. Result ID: 6a1d4034fbc373592c165d8e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### Mahalanobis bACC: 61.1. Result ID: 6a1d4034fbc373592c165d8c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### STAC bACC: 59.1. Result ID: 6a1d4034fbc373592c165d93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### ACE bACC: 59.1. Result ID: 6a1d4034fbc373592c165d94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### EigenScore bACC: 56.6. Result ID: 6a1d4034fbc373592c165d92. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### RND bACC: 55.1. Result ID: 6a1d4034fbc373592c165d8f. [Reporting paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### Cluster Entropy bACC: 52.6. Result ID: 6a1d4034fbc373592c165d91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. ### SAFE-LSTM bACC: 48.4. Result ID: 6a1d4034fbc373592c165d95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy":"π0.5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.