# Question Answering on TruthfulQA (TruthfulQA Metric) Source: [Question Answering on TruthfulQA (TruthfulQA Metric)]() Collection index: [llms.txt]() Data updated: 2026-06-19T08:33:42.653Z Evaluation of model performance on the TruthfulQA dataset using the primary metric designated as 'TruthfulQA Score'. 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 | Question Answering | | Dataset (including any stated split) | TruthfulQA | | Metric | TruthfulQA Score | | Unit | SCALAR | | 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 - [TruthfulQA Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | TruthfulQA Score | Result date | Report paper | | --- | --- | --- | --- | | Attacked | 28.2 | Not recorded | 6a34cd18206542251d7c5f38 | | QVec | 23.8 | Not recorded | 6a34cd18206542251d7c5f38 | | QVec | 23.3 | Not recorded | 6a34cd18206542251d7c5f38 | | Gaussian | 22.4 | Not recorded | 6a34cd18206542251d7c5f38 | | Gaussian | 22.4 | Not recorded | 6a34cd18206542251d7c5f38 | | Gaussian | 22.2 | Not recorded | 6a34cd18206542251d7c5f38 | | QVec | 22 | Not recorded | 6a34cd18206542251d7c5f38 | | Attacked | 21.9 | Not recorded | 6a34cd18206542251d7c5f38 | | Attacked | 21.7 | Not recorded | 6a34cd18206542251d7c5f38 | | Attacked | 21.1 | Not recorded | 6a34cd18206542251d7c5f38 | | Attacked | 20.8 | Not recorded | 6a34cd18206542251d7c5f38 | | Attacked | 20.3 | Not recorded | 6a34cd18206542251d7c5f38 | ### Attacked TruthfulQA Score: 28.2. Result ID: 6a34fec8fbc373592c32619e. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"LLM.int8()"}. This record does not assert independent reproduction. ### QVec TruthfulQA Score: 23.8. Result ID: 6a34fec8fbc373592c3261a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"FP4"}. This record does not assert independent reproduction. ### QVec TruthfulQA Score: 23.3. Result ID: 6a34fec8fbc373592c3261a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"LLM.int8()"}. This record does not assert independent reproduction. ### Gaussian TruthfulQA Score: 22.4. Result ID: 6a34fec8fbc373592c3261a3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"FP4"}. This record does not assert independent reproduction. ### Gaussian TruthfulQA Score: 22.4. Result ID: 6a34fec8fbc373592c3261a7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"NF4"}. This record does not assert independent reproduction. ### Gaussian TruthfulQA Score: 22.2. Result ID: 6a34fec8fbc373592c32619f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"LLM.int8()"}. This record does not assert independent reproduction. ### QVec TruthfulQA Score: 22. Result ID: 6a34fec8fbc373592c3261a8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Precision":"NF4"}. This record does not assert independent reproduction. ### Attacked TruthfulQA Score: 21.9. Result ID: 6a34fec8fbc373592c3261a5. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"FP32"}. This record does not assert independent reproduction. ### Attacked TruthfulQA Score: 21.7. Result ID: 6a34fec8fbc373592c3261a1. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"FP32"}. This record does not assert independent reproduction. ### Attacked TruthfulQA Score: 21.1. Result ID: 6a34fec8fbc373592c32619d. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"FP32"}. This record does not assert independent reproduction. ### Attacked TruthfulQA Score: 20.8. Result ID: 6a34fec8fbc373592c3261a2. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"FP4"}. This record does not assert independent reproduction. ### Attacked TruthfulQA Score: 20.3. Result ID: 6a34fec8fbc373592c3261a6. [Reporting paper]() Recorded evaluation configuration: {"Inference Precision":"NF4"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.