# Natural Language Inference on MNLI Hypothesis-only Bias (in-distribution) Source: [Natural Language Inference on MNLI Hypothesis-only Bias (in-distribution)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:28:05.019Z Evaluation of Natural Language Inference performance specifically on the in-distribution subset of MNLI designed to test hypothesis-only bias. 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) | MNLI Hypothesis-only Bias (in-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 | | --- | --- | --- | --- | | CE | 84.2 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFt | 83.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | PoE | 83.2 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFtTemps | 83.1 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvR | 83.1 | Not recorded | 68fa6b341e209131cea2cdc0 | | PoETemps | 82.9 | Not recorded | 68fa6b341e209131cea2cdc0 | | DRiFtDirichlet | 82.8 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvRTemps | 82.8 | Not recorded | 68fa6b341e209131cea2cdc0 | | PoEDirichlet | 82.7 | Not recorded | 68fa6b341e209131cea2cdc0 | | InvRDirichlet | 82.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMin | 80.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMinTemps | 80.5 | Not recorded | 68fa6b341e209131cea2cdc0 | | LMinDirichlet | 80.1 | Not recorded | 68fa6b341e209131cea2cdc0 | ### CE Accuracy: 84.2. Result ID: 698e7112ffc3901de09d9ec2. [Reporting paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### DRiFt Accuracy: 83.5. Result ID: 698e7112ffc3901de09d9ec6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### PoE Accuracy: 83.2. Result ID: 698e7112ffc3901de09d9ec3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### DRiFtTemps Accuracy: 83.1. Result ID: 698e7112ffc3901de09d9ec7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### InvR Accuracy: 83.1. Result ID: 698e7112ffc3901de09d9ec9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### PoETemps Accuracy: 82.9. Result ID: 698e7112ffc3901de09d9ec4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### DRiFtDirichlet Accuracy: 82.8. Result ID: 698e7112ffc3901de09d9ec8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### InvRTemps Accuracy: 82.8. Result ID: 698e7112ffc3901de09d9eca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### PoEDirichlet Accuracy: 82.7. Result ID: 698e7112ffc3901de09d9ec5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### InvRDirichlet Accuracy: 82.5. Result ID: 698e7112ffc3901de09d9ecb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. ### LMin Accuracy: 80.5. Result ID: 698e7112ffc3901de09d9ecc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"None"}. This record does not assert independent reproduction. ### LMinTemps Accuracy: 80.5. Result ID: 698e7112ffc3901de09d9ecd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Temperature Scaling"}. This record does not assert independent reproduction. ### LMinDirichlet Accuracy: 80.1. Result ID: 698e7112ffc3901de09d9ece. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Calibrator":"Dirichlet"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.