# Truthful Question Answering on TruthfulQA (MC1 & MC2) Source: [Truthful Question Answering on TruthfulQA (MC1 & MC2)]() Collection index: [llms.txt]() Data updated: 2026-05-28T10:32:28.786Z Evaluation of model performance on the TruthfulQA dataset, focusing on the MC1 and MC2 metrics. 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 | Truthful Question Answering | | Dataset (including any stated split) | TruthfulQA | | Metric | MC1 Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 36 of 36 | | 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 - [MC1 Score](): Higher is better; unit: PERCENT. - [MC2 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | MC1 Score | Result date | Report paper | | --- | --- | --- | --- | | MARI | 81.94 | Not recorded | 6a17cc58dbd7150506f81657 | | MARI | 67.93 | Not recorded | 6a17cc58dbd7150506f81657 | | MARI | 66.67 | Not recorded | 6a17cc58dbd7150506f81657 | | MARI | 65.74 | Not recorded | 6a17cc58dbd7150506f81657 | | MARI | 64.35 | Not recorded | 6a17cc58dbd7150506f81657 | | MARI | 61.81 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 59.8 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 55.6 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 52.33 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 50.58 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 50.46 | Not recorded | 6a17cc58dbd7150506f81657 | | ReFT | 49.83 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 47.92 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 46.84 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 46.78 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 44.77 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 43.98 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 42.2 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 40.44 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 39.81 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 36.19 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 36.19 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 35.45 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 35.45 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 33.8 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 33.5 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 32.84 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 32.76 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 32.03 | Not recorded | 6a17cc58dbd7150506f81657 | | NL-ITI | 31.3 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 31.12 | Not recorded | 6a17cc58dbd7150506f81657 | | ITI | 30.81 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 29.64 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 28.7 | Not recorded | 6a17cc58dbd7150506f81657 | | CAA | 28.44 | Not recorded | 6a17cc58dbd7150506f81657 | | Vanilla | 26.75 | Not recorded | 6a17cc58dbd7150506f81657 | ### MARI MC1 Score: 81.94. Result ID: 6a1818f8fbc373592c128cf8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### MARI MC1 Score: 67.93. Result ID: 6a1818f8fbc373592c128cf2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### MARI MC1 Score: 66.67. Result ID: 6a1818f8fbc373592c128ce6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### MARI MC1 Score: 65.74. Result ID: 6a1818f8fbc373592c128cec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. ### MARI MC1 Score: 64.35. Result ID: 6a1818f8fbc373592c128cda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### MARI MC1 Score: 61.81. Result ID: 6a1818f8fbc373592c128ce0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 59.8. Result ID: 6a1818f8fbc373592c128ce5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 55.6. Result ID: 6a1818f8fbc373592c128cf7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 52.33. Result ID: 6a1818f8fbc373592c128cf1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 50.58. Result ID: 6a1818f8fbc373592c128cdf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 50.46. Result ID: 6a1818f8fbc373592c128cd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### ReFT MC1 Score: 49.83. Result ID: 6a1818f8fbc373592c128ceb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 47.92. Result ID: 6a1818f8fbc373592c128cf6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 46.84. Result ID: 6a1818f8fbc373592c128cf0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 46.78. Result ID: 6a1818f8fbc373592c128cf5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 44.77. Result ID: 6a1818f8fbc373592c128cf4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 43.98. Result ID: 6a1818f8fbc373592c128cf3. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 42.2. Result ID: 6a1818f8fbc373592c128cef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 40.44. Result ID: 6a1818f8fbc373592c128cee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 39.81. Result ID: 6a1818f8fbc373592c128ced. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-14B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 36.19. Result ID: 6a1818f8fbc373592c128cde. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 36.19. Result ID: 6a1818f8fbc373592c128ce4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 35.45. Result ID: 6a1818f8fbc373592c128cdd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 35.45. Result ID: 6a1818f8fbc373592c128ce3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 33.8. Result ID: 6a1818f8fbc373592c128cd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 33.5. Result ID: 6a1818f8fbc373592c128ce2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 32.84. Result ID: 6a1818f8fbc373592c128cea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 32.76. Result ID: 6a1818f8fbc373592c128ce1. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-2-13B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 32.03. Result ID: 6a1818f8fbc373592c128cd5. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### NL-ITI MC1 Score: 31.3. Result ID: 6a1818f8fbc373592c128cd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 31.12. Result ID: 6a1818f8fbc373592c128ce9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. ### ITI MC1 Score: 30.81. Result ID: 6a1818f8fbc373592c128cd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 29.64. Result ID: 6a1818f8fbc373592c128cdc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 28.7. Result ID: 6a1818f8fbc373592c128cdb. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-3-8B"}. This record does not assert independent reproduction. ### CAA MC1 Score: 28.44. Result ID: 6a1818f8fbc373592c128ce8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. ### Vanilla MC1 Score: 26.75. Result ID: 6a1818f8fbc373592c128ce7. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.