# Hallucination Detection on TruthfulQA EN Source: [Hallucination Detection on TruthfulQA EN]() Collection index: [llms.txt]() Data updated: 2026-07-07T06:41:55.660Z Evaluation of model performance on detecting hallucinations using the English subset of the TruthfulQA dataset, measured by F1-Score, AUC-ROC, and PR-AUC. 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 | Hallucination Detection | | Dataset (including any stated split) | TruthfulQA EN | | Metric | F1-Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [F1-Score](): Higher is better; unit: PERCENT. - [AUC-ROC](): Higher is better; unit: PERCENT. - [PR-AUC](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1-Score | Result date | Report paper | | --- | --- | --- | --- | | CrossHallu | 88.34 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 84.77 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 83.38 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 82.67 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 81.88 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 79.38 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 79 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 79 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 79 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 78.81 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 78.81 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 77.07 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 76.77 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 76.45 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 76.19 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 59.22 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 57.86 | Not recorded | 6a4c882985dba69fef8b2a32 | | CrossHallu | 6.12 | Not recorded | 6a4c882985dba69fef8b2a32 | ### CrossHallu F1-Score: 88.34. Result ID: 6a4c9fb360b88bb339c6d860. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Aya","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 84.77. Result ID: 6a4c9fb360b88bb339c6d85b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Allam","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 83.38. Result ID: 6a4c9fb360b88bb339c6d861. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Allam","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 82.67. Result ID: 6a4c9fb360b88bb339c6d86a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 81.88. Result ID: 6a4c9fb360b88bb339c6d867. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Allam","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 79.38. Result ID: 6a4c9fb360b88bb339c6d86c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Aya","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 79. Result ID: 6a4c9fb360b88bb339c6d863. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Phi4-mini","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 79. Result ID: 6a4c9fb360b88bb339c6d866. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Aya","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 79. Result ID: 6a4c9fb360b88bb339c6d869. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Phi4-mini","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 78.81. Result ID: 6a4c9fb360b88bb339c6d85d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Phi4-mini","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 78.81. Result ID: 6a4c9fb360b88bb339c6d85e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 77.07. Result ID: 6a4c9fb360b88bb339c6d868. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Silma","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 76.77. Result ID: 6a4c9fb360b88bb339c6d862. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Silma","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 76.45. Result ID: 6a4c9fb360b88bb339c6d85c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Silma","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 76.19. Result ID: 6a4c9fb360b88bb339c6d864. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 59.22. Result ID: 6a4c9fb360b88bb339c6d86b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5","Train dataset":"HalluScore","Transfer setting group":"Cross-lingual + cross-domain (CL-CD)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 57.86. Result ID: 6a4c9fb360b88bb339c6d85f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5","Train dataset":"TruthfulQA EN","Transfer setting group":"Same-language baseline (BS)"}. This record does not assert independent reproduction. ### CrossHallu F1-Score: 6.12. Result ID: 6a4c9fb360b88bb339c6d865. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5","Train dataset":"TruthfulQA AR","Transfer setting group":"Cross-lingual (CL)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.