# Multiple-choice factual recall on POLYFACT High Source: [Multiple-choice factual recall on POLYFACT High]() Collection index: [llms.txt]() Data updated: 2026-06-08T06:32:02.694Z Evaluation of model performance on the POLYFACT High dataset focusing on multiple-choice factual recall questions, measured using 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 | Multiple-choice factual recall | | Dataset (including any stated split) | POLYFACT High | | Metric | Accuracy | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | GRPO | 73.15 | Not recorded | 6a264c8590966ade919be956 | | Baseline | 66.69 | Not recorded | 6a264c8590966ade919be956 | | GRPO | 64.21 | Not recorded | 6a264c8590966ade919be956 | | CPT + GRPO | 64.11 | Not recorded | 6a264c8590966ade919be956 | | SFT | 63.25 | Not recorded | 6a264c8590966ade919be956 | | CPT | 61.29 | Not recorded | 6a264c8590966ade919be956 | | CPT + GRPO | 61.26 | Not recorded | 6a264c8590966ade919be956 | | CPT + SFT | 59.2 | Not recorded | 6a264c8590966ade919be956 | | CPT + SFT | 58.1 | Not recorded | 6a264c8590966ade919be956 | | Baseline | 57.93 | Not recorded | 6a264c8590966ade919be956 | | CPT | 57.89 | Not recorded | 6a264c8590966ade919be956 | | SFT | 56.33 | Not recorded | 6a264c8590966ade919be956 | ### GRPO Accuracy: 73.15. Result ID: 6a266179fbc373592c226c36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### Baseline Accuracy: 66.69. Result ID: 6a266179fbc373592c226c32. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### GRPO Accuracy: 64.21. Result ID: 6a266179fbc373592c226c30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. ### CPT + GRPO Accuracy: 64.11. Result ID: 6a266179fbc373592c226c37. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### SFT Accuracy: 63.25. Result ID: 6a266179fbc373592c226c34. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### CPT Accuracy: 61.29. Result ID: 6a266179fbc373592c226c33. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### CPT + GRPO Accuracy: 61.26. Result ID: 6a266179fbc373592c226c31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. ### CPT + SFT Accuracy: 59.2. Result ID: 6a266179fbc373592c226c2f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. ### CPT + SFT Accuracy: 58.1. Result ID: 6a266179fbc373592c226c35. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction. ### Baseline Accuracy: 57.93. Result ID: 6a266179fbc373592c226c2c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. ### CPT Accuracy: 57.89. Result ID: 6a266179fbc373592c226c2d. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. ### SFT Accuracy: 56.33. Result ID: 6a266179fbc373592c226c2e. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.