# Question Answering on ARC Challenge (PR-AUC) Source: [Question Answering on ARC Challenge (PR-AUC)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:55:16.370Z Evaluation of model performance on the ARC Challenge dataset for Question Answering, specifically measured using the Precision-Recall Area Under Curve (PR-AUC) metric. 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 | Question Answering | | Dataset (including any stated split) | ARC Challenge | | Metric | PR-AUC | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [PR-AUC](): Higher is better; unit: SCALAR. ## Recorded results | Method | PR-AUC | Result date | Report paper | | --- | --- | --- | --- | | Eccentricity | 0.636 | Not recorded | 693a4ff46cfd488761a54954 | | EigVecDissimilarity | 0.621 | Not recorded | 693a4ff46cfd488761a54954 | | CocoaPPL | 0.61 | Not recorded | 693a4ff46cfd488761a54954 | | CocoaMSP | 0.606 | Not recorded | 693a4ff46cfd488761a54954 | | Dissimilarity | 0.604 | Not recorded | 693a4ff46cfd488761a54954 | | CocoaMSP | 0.595 | Not recorded | 693a4ff46cfd488761a54954 | | CocoaPPL | 0.595 | Not recorded | 693a4ff46cfd488761a54954 | | EigVecDissimilarity | 0.562 | Not recorded | 693a4ff46cfd488761a54954 | | Eccentricity | 0.549 | Not recorded | 693a4ff46cfd488761a54954 | | Dissimilarity | 0.545 | Not recorded | 693a4ff46cfd488761a54954 | | Degree Matrix | 0.473 | Not recorded | 693a4ff46cfd488761a54954 | | Semantic Entropy | 0.454 | Not recorded | 693a4ff46cfd488761a54954 | | Semantic Entropy | 0.431 | Not recorded | 693a4ff46cfd488761a54954 | | Degree Matrix | 0.419 | Not recorded | 693a4ff46cfd488761a54954 | | SAR | 0.269 | Not recorded | 693a4ff46cfd488761a54954 | | SAR | 0.264 | Not recorded | 693a4ff46cfd488761a54954 | ### Eccentricity PR-AUC: 0.636. Result ID: 699a36fa5698665ec4bd1cd2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### EigVecDissimilarity PR-AUC: 0.621. Result ID: 699a36fa5698665ec4bd1cd4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### CocoaPPL PR-AUC: 0.61. Result ID: 699a36fa5698665ec4bd1cd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### CocoaMSP PR-AUC: 0.606. Result ID: 699a36fa5698665ec4bd1cd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Dissimilarity PR-AUC: 0.604. Result ID: 699a36fa5698665ec4bd1cd0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### CocoaMSP PR-AUC: 0.595. Result ID: 699a36fa5698665ec4bd1cd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### CocoaPPL PR-AUC: 0.595. Result ID: 699a36fa5698665ec4bd1cd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### EigVecDissimilarity PR-AUC: 0.562. Result ID: 699a36fa5698665ec4bd1cd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Eccentricity PR-AUC: 0.549. Result ID: 699a36fa5698665ec4bd1cd1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Dissimilarity PR-AUC: 0.545. Result ID: 699a36fa5698665ec4bd1ccf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Degree Matrix PR-AUC: 0.473. Result ID: 699a36fa5698665ec4bd1ccc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Semantic Entropy PR-AUC: 0.454. Result ID: 699a36fa5698665ec4bd1cca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Semantic Entropy PR-AUC: 0.431. Result ID: 699a36fa5698665ec4bd1cc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### Degree Matrix PR-AUC: 0.419. Result ID: 699a36fa5698665ec4bd1ccb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### SAR PR-AUC: 0.269. Result ID: 699a36fa5698665ec4bd1cce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"beam-search","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. ### SAR PR-AUC: 0.264. Result ID: 699a36fa5698665ec4bd1ccd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"multinomial-sampling","Backbone":"Gemma 3 4B base"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.