# Information Retrieval on SciFact (Faith., Hit@k, p95) Source: [Information Retrieval on SciFact (Faith., Hit@k, p95)]() Collection index: [llms.txt]() Data updated: 2026-03-31T08:40:30.452Z Evaluation of Information Retrieval performance on the SciFact dataset, focusing on faithfulness, ranking accuracy (Hit@k), and latency (p95). 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 | Information Retrieval | | Dataset (including any stated split) | SciFact | | Metric | Faithfulness | | 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 - [Faithfulness](): Higher is better; unit: PERCENT. - [Hit@k](): Higher is better; unit: PERCENT. - [Latency (p95)](): Lower is better; unit: TIME. ## Recorded results | Method | Faithfulness | Result date | Report paper | | --- | --- | --- | --- | | gpt-4.1-mini | 67 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1-mini | 63 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1-mini | 63 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 62 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 61 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 60 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 59 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1-mini | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1-mini | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1-mini | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 58 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 56 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-5.2 | 56 | Not recorded | 69cb54fd0636e5bf4afc9e72 | | gpt-4.1 | 55 | Not recorded | 69cb54fd0636e5bf4afc9e72 | ### gpt-4.1-mini Faithfulness: 67. Result ID: 69cb882f47f34ce924b3dbc5. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1-mini Faithfulness: 63. Result ID: 69cb882f47f34ce924b3dbc3. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1-mini Faithfulness: 63. Result ID: 69cb882f47f34ce924b3dbc7. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 62. Result ID: 69cb882f47f34ce924b3dbc8. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 61. Result ID: 69cb882f47f34ce924b3dbcc. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 60. Result ID: 69cb882f47f34ce924b3dbca. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 59. Result ID: 69cb882f47f34ce924b3dbbf. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbbc. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbbd. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbc1. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1-mini Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbc2. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1-mini Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbc4. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1-mini Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbc6. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbc9. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Cost-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 58. Result ID: 69cb882f47f34ce924b3dbcd. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 56. Result ID: 69cb882f47f34ce924b3dbbe. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. ### gpt-5.2 Faithfulness: 56. Result ID: 69cb882f47f34ce924b3dbcb. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"Risk-first","k":"5","Runs":"3"}. This record does not assert independent reproduction. ### gpt-4.1 Faithfulness: 55. Result ID: 69cb882f47f34ce924b3dbc0. [Reporting paper]() Recorded evaluation configuration: {"Scenario":"SLA-first","k":"3","Runs":"3"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.