# Multilingual Factual Reasoning on mTruthfulQA Source: [Multilingual Factual Reasoning on mTruthfulQA]() Collection index: [llms.txt]() Data updated: 2026-04-15T07:12:08.752Z Evaluation of model performance on the multilingual version of TruthfulQA (mTruthfulQA), focusing on factual reasoning capabilities across numerous languages. 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 | Multilingual Factual Reasoning | | Dataset (including any stated split) | mTruthfulQA | | Metric | Score (ar) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [Score (ar)](): Higher is better; unit: PERCENT. - [Score (bn)](): Higher is better; unit: PERCENT. - [Score (ca)](): Higher is better; unit: PERCENT. - [Score (da)](): Higher is better; unit: PERCENT. - [Score (de)](): Higher is better; unit: PERCENT. - [Score (en)](): Higher is better; unit: PERCENT. - [Score (es)](): Higher is better; unit: PERCENT. - [Score (eu)](): Higher is better; unit: PERCENT. - [Score (fr)](): Higher is better; unit: PERCENT. - [Score (gu)](): Higher is better; unit: PERCENT. - [Score (hi)](): Higher is better; unit: PERCENT. - [Score (hr)](): Higher is better; unit: PERCENT. - [Score (hu)](): Higher is better; unit: PERCENT. - [Score (hy)](): Higher is better; unit: PERCENT. - [Score (id)](): Higher is better; unit: PERCENT. - [Score (it)](): Higher is better; unit: PERCENT. - [Score (kn)](): Higher is better; unit: PERCENT. - [Score (ml)](): Higher is better; unit: PERCENT. - [Score (mr)](): Higher is better; unit: PERCENT. - [Score (ne)](): Higher is better; unit: PERCENT. - [Score (nl)](): Higher is better; unit: PERCENT. - [Score (pt)](): Higher is better; unit: PERCENT. - [Score (ro)](): Higher is better; unit: PERCENT. - [Score (ru)](): Higher is better; unit: PERCENT. - [Score (sk)](): Higher is better; unit: PERCENT. - [Score (sr)](): Higher is better; unit: PERCENT. - [Score (sv)](): Higher is better; unit: PERCENT. - [Score (ta)](): Higher is better; unit: PERCENT. - [Score (te)](): Higher is better; unit: PERCENT. - [Score (uk)](): Higher is better; unit: PERCENT. - [Score (vi)](): Higher is better; unit: PERCENT. - [Score (zh)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Score (ar) | Result date | Report paper | | --- | --- | --- | --- | | LLaMA3.1-8B-AdaMCOT | 54.46 | Not recorded | 68fa7a601e209131cea82b5f | | Qwen2.5-7B | 52.78 | Not recorded | 68fa7a601e209131cea82b5f | | Qwen2.5-7B-AdaMCOT | 47.35 | Not recorded | 68fa7a601e209131cea82b5f | | LLaMA3.1-8B | 46.96 | Not recorded | 68fa7a601e209131cea82b5f | | Qwen2.5-7B-QAlign | 45.8 | Not recorded | 68fa7a601e209131cea82b5f | | Qwen2.5-7B-AutoCAP | 44.18 | Not recorded | 68fa7a601e209131cea82b5f | | LLaMA3.1-8B-QAlign | 43.47 | Not recorded | 68fa7a601e209131cea82b5f | | LLaMA3.1-8B-AutoCAP | 41.01 | Not recorded | 68fa7a601e209131cea82b5f | ### LLaMA3.1-8B-AdaMCOT Score (ar): 54.46. Result ID: 69df390f47f34ce924c78612. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaMA3.1-8B","Method":"AdaMCOT"}. This record does not assert independent reproduction. ### Qwen2.5-7B Score (ar): 52.78. Result ID: 69df390f47f34ce924c78613. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### Qwen2.5-7B-AdaMCOT Score (ar): 47.35. Result ID: 69df390f47f34ce924c78616. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B","Method":"AdaMCOT"}. This record does not assert independent reproduction. ### LLaMA3.1-8B Score (ar): 46.96. Result ID: 69df390f47f34ce924c7860f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaMA3.1-8B"}. This record does not assert independent reproduction. ### Qwen2.5-7B-QAlign Score (ar): 45.8. Result ID: 69df390f47f34ce924c78615. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B","Method":"QAlign"}. This record does not assert independent reproduction. ### Qwen2.5-7B-AutoCAP Score (ar): 44.18. Result ID: 69df390f47f34ce924c78614. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B","Method":"AutoCAP"}. This record does not assert independent reproduction. ### LLaMA3.1-8B-QAlign Score (ar): 43.47. Result ID: 69df390f47f34ce924c78611. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaMA3.1-8B","Method":"QAlign"}. This record does not assert independent reproduction. ### LLaMA3.1-8B-AutoCAP Score (ar): 41.01. Result ID: 69df390f47f34ce924c78610. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaMA3.1-8B","Method":"AutoCAP"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.