# Generative Question Answering on MedHallu Source: [Generative Question Answering on MedHallu]() Collection index: [llms.txt]() Data updated: 2026-06-02T09:36:20.409Z Evaluation of model performance on the MedHallu dataset for Generative Question Answering, measured using HALL, ABS, FACT, and F-score metrics. 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 | Generative Question Answering | | Dataset (including any stated split) | MedHallu | | Metric | HALL Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [HALL Score](): Higher is better; unit: PERCENT. - [ABS Score](): Higher is better; unit: PERCENT. - [FACT Score](): Higher is better; unit: PERCENT. - [F-score](): Higher is better; unit: PERCENT. ## Recorded results | Method | HALL Score | Result date | Report paper | | --- | --- | --- | --- | | Base | 67.33 | Not recorded | 6a1e6393dbd7150506f85222 | | Append-idk | 66.67 | Not recorded | 6a1e6393dbd7150506f85222 | | Plan-and-Solve | 65 | Not recorded | 6a1e6393dbd7150506f85222 | | RCSP | 64.33 | Not recorded | 6a1e6393dbd7150506f85222 | | Base | 62 | Not recorded | 6a1e6393dbd7150506f85222 | | Plan-and-Solve | 61.67 | Not recorded | 6a1e6393dbd7150506f85222 | | EmoPrompt (avg) | 60.97 | Not recorded | 6a1e6393dbd7150506f85222 | | EmoPrompt (avg) | 59.64 | Not recorded | 6a1e6393dbd7150506f85222 | | RCSP | 57.67 | Not recorded | 6a1e6393dbd7150506f85222 | | Append-idk | 47.33 | Not recorded | 6a1e6393dbd7150506f85222 | ### Base HALL Score: 67.33. Result ID: 6a1ea334fbc373592c191fac. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction. ### Append-idk HALL Score: 66.67. Result ID: 6a1ea334fbc373592c191f94. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction. ### Plan-and-Solve HALL Score: 65. Result ID: 6a1ea334fbc373592c191faf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction. ### RCSP HALL Score: 64.33. Result ID: 6a1ea334fbc373592c191fb0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction. ### Base HALL Score: 62. Result ID: 6a1ea334fbc373592c191f93. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction. ### Plan-and-Solve HALL Score: 61.67. Result ID: 6a1ea334fbc373592c191f96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction. ### EmoPrompt (avg) HALL Score: 60.97. Result ID: 6a1ea334fbc373592c191fae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction. ### EmoPrompt (avg) HALL Score: 59.64. Result ID: 6a1ea334fbc373592c191f95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction. ### RCSP HALL Score: 57.67. Result ID: 6a1ea334fbc373592c191f97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction. ### Append-idk HALL Score: 47.33. Result ID: 6a1ea334fbc373592c191fad. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.