# Truthful Question Answering on TruthfulQA f\_con o=5 Source: [Truthful Question Answering on TruthfulQA f\_con o=5]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:26:50.617Z Evaluation of model performance on the TruthfulQA dataset using the 'f\_con o=5' configuration, reporting Accuracy and Residual Contamination (RC) metrics across Exact, Semantic-level, and Domain-level comparisons. 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 f\_con o=5 | | Metric | Acc (Exact) | | 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 - [Acc (Exact)](): Higher is better; unit: PERCENT. - [RC (Exact)](): Lower is better; unit: PERCENT. - [Acc (Semantic-level)](): Higher is better; unit: PERCENT. - [RC (Semantic-level)](): Lower is better; unit: PERCENT. - [Acc (Domain-level)](): Higher is better; unit: PERCENT. - [RC (Domain-level)](): Lower is better; unit: PERCENT. ## Recorded results | Method | Acc (Exact) | Result date | Report paper | | --- | --- | --- | --- | | W/O Decontamination | 100 | Not recorded | 6979980ea2b28e761d87cbce | | ITD | 100 | Not recorded | 6979980ea2b28e761d87cbce | | W/O Decontamination | 99.5 | Not recorded | 6979980ea2b28e761d87cbce | | ITD | 99.5 | Not recorded | 6979980ea2b28e761d87cbce | | W/O Decontamination | 99.5 | Not recorded | 6979980ea2b28e761d87cbce | | Short Circuit | 91.9 | Not recorded | 6979980ea2b28e761d87cbce | | ITD | 90.9 | Not recorded | 6979980ea2b28e761d87cbce | | TED | 86.6 | Not recorded | 6979980ea2b28e761d87cbce | | Shortcut Neuron | 85.2 | Not recorded | 6979980ea2b28e761d87cbce | | DeconIEP | 79 | Not recorded | 6979980ea2b28e761d87cbce | | DeconIEP | 66 | Not recorded | 6979980ea2b28e761d87cbce | | Short Circuit | 65 | Not recorded | 6979980ea2b28e761d87cbce | | Shortcut Neuron | 62.1 | Not recorded | 6979980ea2b28e761d87cbce | | DeconIEP | 59.5 | Not recorded | 6979980ea2b28e761d87cbce | | Short Circuit | 58.9 | Not recorded | 6979980ea2b28e761d87cbce | | Shortcut Neuron | 51.7 | Not recorded | 6979980ea2b28e761d87cbce | | TED | 19.1 | Not recorded | 6979980ea2b28e761d87cbce | | TED | 8.6 | Not recorded | 6979980ea2b28e761d87cbce | ### W/O Decontamination Acc (Exact): 100. Result ID: 699669b45698665ec4b04a5f. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"–"}. This record does not assert independent reproduction. ### ITD Acc (Exact): 100. Result ID: 699669b45698665ec4b04a61. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"Black-box"}. This record does not assert independent reproduction. ### W/O Decontamination Acc (Exact): 99.5. Result ID: 699669b45698665ec4b04a59. [Reporting paper]() Recorded evaluation configuration: {"Model":"Mistral","Access":"–"}. This record does not assert independent reproduction. ### ITD Acc (Exact): 99.5. Result ID: 699669b45698665ec4b04a5b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mistral","Access":"Black-box"}. This record does not assert independent reproduction. ### W/O Decontamination Acc (Exact): 99.5. Result ID: 699669b45698665ec4b04a65. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"–"}. This record does not assert independent reproduction. ### Short Circuit Acc (Exact): 91.9. Result ID: 699669b45698665ec4b04a5d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mistral","Access":"White-box"}. This record does not assert independent reproduction. ### ITD Acc (Exact): 90.9. Result ID: 699669b45698665ec4b04a67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"Black-box"}. This record does not assert independent reproduction. ### TED Acc (Exact): 86.6. Result ID: 699669b45698665ec4b04a66. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"Black-box"}. This record does not assert independent reproduction. ### Shortcut Neuron Acc (Exact): 85.2. Result ID: 699669b45698665ec4b04a68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"White-box"}. This record does not assert independent reproduction. ### DeconIEP Acc (Exact): 79. Result ID: 699669b45698665ec4b04a64. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"White-box"}. This record does not assert independent reproduction. ### DeconIEP Acc (Exact): 66. Result ID: 699669b45698665ec4b04a6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"White-box"}. This record does not assert independent reproduction. ### Short Circuit Acc (Exact): 65. Result ID: 699669b45698665ec4b04a63. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"White-box"}. This record does not assert independent reproduction. ### Shortcut Neuron Acc (Exact): 62.1. Result ID: 699669b45698665ec4b04a5c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mistral","Access":"White-box"}. This record does not assert independent reproduction. ### DeconIEP Acc (Exact): 59.5. Result ID: 699669b45698665ec4b04a5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mistral","Access":"White-box"}. This record does not assert independent reproduction. ### Short Circuit Acc (Exact): 58.9. Result ID: 699669b45698665ec4b04a69. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-3","Access":"White-box"}. This record does not assert independent reproduction. ### Shortcut Neuron Acc (Exact): 51.7. Result ID: 699669b45698665ec4b04a62. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"White-box"}. This record does not assert independent reproduction. ### TED Acc (Exact): 19.1. Result ID: 699669b45698665ec4b04a60. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen 2.5","Access":"Black-box"}. This record does not assert independent reproduction. ### TED Acc (Exact): 8.6. Result ID: 699669b45698665ec4b04a5a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Mistral","Access":"Black-box"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.