# Factuality Detection on TriviaQA, HotpotQA, and CSQA Source: [Factuality Detection on TriviaQA, HotpotQA, and CSQA]() Collection index: [llms.txt]() Data updated: 2026-03-18T08:51:59.545Z Evaluation of model factuality detection capabilities across multiple question-answering datasets, including TriviaQA, HotpotQA, and CSQA, measured using metrics like Average AUROC. 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 | Factuality Detection | | Dataset (including any stated split) | Multiple TriviaQA, HotpotQA, CSQA | | Metric | Average AUROC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 4 of 4 | | 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 - [Average AUROC](): Higher is better; unit: PERCENT. ## Recorded results | Method | Average AUROC | Result date | Report paper | | --- | --- | --- | --- | | DynHD | 72.9 | Not recorded | 69ba318b0636e5bf4afc0802 | | TraceDet | 66.3 | Not recorded | 69ba318b0636e5bf4afc0802 | | TSV | 61.3 | Not recorded | 69ba318b0636e5bf4afc0802 | | CCS | 58.2 | Not recorded | 69ba318b0636e5bf4afc0802 | ### DynHD Average AUROC: 72.9. Result ID: 69ba66e347f34ce924a46655. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Trajectory-based","Evaluation Protocol":"Zero-shot cross-task generalization"}. This record does not assert independent reproduction. ### TraceDet Average AUROC: 66.3. Result ID: 69ba66e347f34ce924a46654. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Trajectory-based","Evaluation Protocol":"Zero-shot cross-task generalization"}. This record does not assert independent reproduction. ### TSV Average AUROC: 61.3. Result ID: 69ba66e347f34ce924a46653. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Latent-based","Evaluation Protocol":"Zero-shot cross-task generalization"}. This record does not assert independent reproduction. ### CCS Average AUROC: 58.2. Result ID: 69ba66e347f34ce924a46652. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Latent-based","Evaluation Protocol":"Zero-shot cross-task generalization"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.