# Medical Visual Question Answering on SLAKE 12 Source: [Medical Visual Question Answering on SLAKE 12]() Collection index: [llms.txt]() Data updated: 2026-06-19T08:49:06.485Z Evaluation of model performance on the SLAKE 12 dataset for Medical Visual Question Answering, reporting F1, Open Recall, LLM Open, and Closed Accuracy. 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 | Medical Visual Question Answering | | Dataset (including any stated split) | Slake 12 | | Metric | F1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 - [F1](): Higher is better; unit: PERCENT. - [Open Recall](): Higher is better; unit: PERCENT. - [LLM Open Score](): Higher is better; unit: PERCENT. - [Closed Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1 | Result date | Report paper | | --- | --- | --- | --- | | RadGrounder | 88.4 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 87.7 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 87.6 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 87.2 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 86.9 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 86.4 | Not recorded | 6a34cd57206542251d7c61fa | | BiomedGPT-B | 85.2 | Not recorded | 6a34cd57206542251d7c61fa | | RadFM (14B) | 84.4 | Not recorded | 6a34cd57206542251d7c61fa | | Med-Gemini | 75.8 | Not recorded | 6a34cd57206542251d7c61fa | | MedG. (4B) | 72.3 | Not recorded | 6a34cd57206542251d7c61fa | | Gemma 3 (4B) | 40.2 | Not recorded | 6a34cd57206542251d7c61fa | | PaliG. 2 (3B) | 24.5 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 19.2 | Not recorded | 6a34cd57206542251d7c61fa | | RadGrounder | 18.3 | Not recorded | 6a34cd57206542251d7c61fa | ### RadGrounder F1: 88.4. Result ID: 6a350275fbc373592c328411. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Slake, VQA-RAD, Detect.","Vision Encoder":"Unfreeze finet. SL."}. This record does not assert independent reproduction. ### RadGrounder F1: 87.7. Result ID: 6a350275fbc373592c328410. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Slake, VQA-RAD, Detect.","Vision Encoder":"Frozen SigLIP (adapted)"}. This record does not assert independent reproduction. ### RadGrounder F1: 87.6. Result ID: 6a350275fbc373592c328412. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Slake, VQA-RAD, Detect.","Vision Encoder":"Unfreeze orig. SL."}. This record does not assert independent reproduction. ### RadGrounder F1: 87.2. Result ID: 6a350275fbc373592c328416. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Slake, VQA-RAD"}. This record does not assert independent reproduction. ### RadGrounder F1: 86.9. Result ID: 6a350275fbc373592c328415. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Detect."}. This record does not assert independent reproduction. ### RadGrounder F1: 86.4. Result ID: 6a350275fbc373592c328413. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Slake, VQA-RAD, Detect."}. This record does not assert independent reproduction. ### BiomedGPT-B F1: 85.2. Result ID: 6a350275fbc373592c32840c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RadFM (14B) F1: 84.4. Result ID: 6a350275fbc373592c32840e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"14B"}. This record does not assert independent reproduction. ### Med-Gemini F1: 75.8. Result ID: 6a350275fbc373592c32840f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MedG. (4B) F1: 72.3. Result ID: 6a350275fbc373592c32840b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction. ### Gemma 3 (4B) F1: 40.2. Result ID: 6a350275fbc373592c32840a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction. ### PaliG. 2 (3B) F1: 24.5. Result ID: 6a350275fbc373592c328409. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"3B"}. This record does not assert independent reproduction. ### RadGrounder F1: 19.2. Result ID: 6a350275fbc373592c328414. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA"}. This record does not assert independent reproduction. ### RadGrounder F1: 18.3. Result ID: 6a350275fbc373592c328417. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"Report, VQA, Segment"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.