# 3D Medical Visual Question Answering on M3D-VQA Source: [3D Medical Visual Question Answering on M3D-VQA]() Collection index: [llms.txt]() Data updated: 2026-06-11T10:28:51.435Z Evaluation of model performance on the M3D-VQA dataset for 3D Medical Visual Question Answering, reporting scores across Plane, Phase, Organ, Abnormality, Location, and Mean 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 | 3D Medical Visual Question Answering | | Dataset (including any stated split) | M3D-VQA | | Metric | Accuracy (Plane) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 - [Accuracy (Plane)](): Higher is better; unit: PERCENT. - [Accuracy (Phase)](): Higher is better; unit: PERCENT. - [Accuracy (Organ)](): Higher is better; unit: PERCENT. - [Accuracy (Abnormality)](): Higher is better; unit: PERCENT. - [Accuracy (Location)](): Higher is better; unit: PERCENT. - [Mean Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Plane) | Result date | Report paper | | --- | --- | --- | --- | | UniReason-Med | 99 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | M3D-LaMed-Llama-2-7B | 98.8 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | M3D-LaMed-Phi-3-4B | 98.6 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | Lingshu-7B | 97.3 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | Qwen2.5VL-7B | 91.3 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | Qwen2.5VL-32B | 90.8 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | MedGemma-4B-IT | 75.5 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | LLaVA-Med-7B | 73.4 | Not recorded | 6a2a40fcd33dfb23e78391f6 | | RadFM | 19.7 | Not recorded | 6a2a40fcd33dfb23e78391f6 | ### UniReason-Med Accuracy (Plane): 99. Result ID: 6a2a8dd2fbc373592c2867bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### M3D-LaMed-Llama-2-7B Accuracy (Plane): 98.8. Result ID: 6a2a8dd2fbc373592c2867bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-2","Parameters":"7B"}. This record does not assert independent reproduction. ### M3D-LaMed-Phi-3-4B Accuracy (Plane): 98.6. Result ID: 6a2a8dd2fbc373592c2867bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Phi-3","Parameters":"4B"}. This record does not assert independent reproduction. ### Lingshu-7B Accuracy (Plane): 97.3. Result ID: 6a2a8dd2fbc373592c2867be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Qwen2.5VL-7B Accuracy (Plane): 91.3. Result ID: 6a2a8dd2fbc373592c2867b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Qwen2.5VL-32B Accuracy (Plane): 90.8. Result ID: 6a2a8dd2fbc373592c2867b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"32B"}. This record does not assert independent reproduction. ### MedGemma-4B-IT Accuracy (Plane): 75.5. Result ID: 6a2a8dd2fbc373592c2867ba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B","Configuration":"Instruction-Tuned"}. This record does not assert independent reproduction. ### LLaVA-Med-7B Accuracy (Plane): 73.4. Result ID: 6a2a8dd2fbc373592c2867bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### RadFM Accuracy (Plane): 19.7. Result ID: 6a2a8dd2fbc373592c2867b7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.