# Natural Language Inference on HANS Syntactic Bias (out-of-distribution) Source: [Natural Language Inference on HANS Syntactic Bias (out-of-distribution)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:28:04.995Z Evaluation of Natural Language Inference performance specifically targeting syntactic biases using the HANS (Heuristic Analysis for NLI Systems) out-of-distribution test set, 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 | Natural Language Inference | | Dataset (including any stated split) | HANS Syntactic Bias (out-of-distribution) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 | | --- | --- | --- | --- | | PoEDirichlet | 70.7 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFtDirichlet | 69.8 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFtTemps | 69.7 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvRTemps | 69.4 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvRDirichlet | 69.4 | Not recorded | 68fa6b341e209131cea2cdc0 | | PoETemps | 69.1 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvR | 68.4 | Not recorded | 68fa6b341e209131cea2cdc0 | | PoE | 68.1 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFt | 66.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMin | 65.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMinTemps | 63.2 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMinDirichlet | 62.7 | Not recorded | 68fa6b341e209131cea2cdc0 | | CE | 61.2 | Not recorded | 68fa6b341e209131cea2cdc0 | ### PoEDirichlet Accuracy: 70.7. Result ID: 698e7112ffc3901de09d9eb8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### DRiFtDirichlet Accuracy: 69.8. Result ID: 698e7112ffc3901de09d9ebb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### DRiFtTemps Accuracy: 69.7. Result ID: 698e7112ffc3901de09d9eba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### InvRTemps Accuracy: 69.4. Result ID: 698e7112ffc3901de09d9ebd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### InvRDirichlet Accuracy: 69.4. Result ID: 698e7112ffc3901de09d9ebe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### PoETemps Accuracy: 69.1. Result ID: 698e7112ffc3901de09d9eb7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### InvR Accuracy: 68.4. Result ID: 698e7112ffc3901de09d9ebc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### PoE Accuracy: 68.1. Result ID: 698e7112ffc3901de09d9eb6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### DRiFt Accuracy: 66.5. Result ID: 698e7112ffc3901de09d9eb9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### LMin Accuracy: 65.5. Result ID: 698e7112ffc3901de09d9ebf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### LMinTemps Accuracy: 63.2. Result ID: 698e7112ffc3901de09d9ec0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### LMinDirichlet Accuracy: 62.7. Result ID: 698e7112ffc3901de09d9ec1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### CE Accuracy: 61.2. Result ID: 698e7112ffc3901de09d9eb5. [Reporting paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.