# Commonsense Anomaly Detection on Urbach and Kutas Commonsense Source: [Commonsense Anomaly Detection on Urbach and Kutas Commonsense]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:27:54.208Z Evaluation of Commonsense Anomaly Detection performance using the Urbach and Kutas Commonsense dataset, measured by 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 | Commonsense Anomaly Detection | | Dataset (including any stated split) | Urbach and Kutas Commonsense | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Masked language model | 93.9 | Not recorded | 68fa69f51e209131cea2541d | | Masked language model | 92.4 | Not recorded | 68fa69f51e209131cea2541d | | Masked language model | 71.2 | Not recorded | 68fa69f51e209131cea2541d | | Gaussian anomaly model | 50 | Not recorded | 68fa69f51e209131cea2541d | | Gaussian anomaly model | 48.5 | Not recorded | 68fa69f51e209131cea2541d | | Gaussian anomaly model | 47 | Not recorded | 68fa69f51e209131cea2541d | ### Masked language model Accuracy: 93.9. Result ID: 698dfb8effc3901de099ea36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa","Anomaly Model":"MLM","Layer Selection":"Last layer"}. This record does not assert independent reproduction. ### Masked language model Accuracy: 92.4. Result ID: 698dfb8effc3901de099ea34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT","Anomaly Model":"MLM","Layer Selection":"Last layer"}. This record does not assert independent reproduction. ### Masked language model Accuracy: 71.2. Result ID: 698dfb8effc3901de099ea38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"XLNet","Anomaly Model":"MLM","Layer Selection":"Last layer"}. This record does not assert independent reproduction. ### Gaussian anomaly model Accuracy: 50. Result ID: 698dfb8effc3901de099ea37. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"XLNet","Anomaly Model":"GM","Layer Selection":"Best layer"}. This record does not assert independent reproduction. ### Gaussian anomaly model Accuracy: 48.5. Result ID: 698dfb8effc3901de099ea35. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"RoBERTa","Anomaly Model":"GM","Layer Selection":"Best layer"}. This record does not assert independent reproduction. ### Gaussian anomaly model Accuracy: 47. Result ID: 698dfb8effc3901de099ea33. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT","Anomaly Model":"GM","Layer Selection":"Best layer"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.