# Medical Visual Question Answering on VQA-RAD Source: [Medical Visual Question Answering on VQA-RAD]() Collection index: [llms.txt]() Data updated: 2026-07-01T08:15:32.479Z Evaluation of model performance on the VQA-RAD dataset for medical visual question answering, reporting Open and Closed accuracy 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 | Medical Visual Question Answering | | Dataset (including any stated split) | VQA-RAD | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 349 of 349 | | 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](): Higher is better; unit: PERCENT. - [Open Accuracy](): Higher is better; unit: PERCENT. - [Closed Accuracy](): Higher is better; unit: PERCENT. - [Recall (Open)](): Higher is better; unit: PERCENT. - [Overall Accuracy](): Higher is better; unit: PERCENT. - [Token-F1 (Open)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Robust-MMR | 80.4 | Not recorded | 699bdf1909c33900fcd804ee | | OctoMed-7B | 79 | Not recorded | 69a66c1a2d603e6de3e3ce72 | | Ours (7B) | 79 | Not recorded | 69a66c1a2d603e6de3e3ce72 | | MedGemma 4B | 78.55 | Not recorded | 68fa7aa61e209131cea84677 | | MedGemma 4B | 78.55 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | KG-CMI | 78.21 | Not recorded | 69cdf8430636e5bf4afcccf3 | | CPRD | 78.2 | Not recorded | 699bdf1909c33900fcd804ee | | LaPA | 78.15 | Not recorded | 69cdf8430636e5bf4afcccf3 | | MedVLSynther-7B | 77.57 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | Ours-7B | 77.16 | Not recorded | 69f190b655a3c32af89fbc0e | | MEVF | 77 | Not recorded | 699bdf1909c33900fcd804ee | | MedVLThinker-32B RL | 76.96 | Not recorded | 68fa7aa61e209131cea84677 | | M3AE | 76.72 | Not recorded | 69cdf8430636e5bf4afcccf3 | | VG-CALF | 76.16 | Not recorded | 69cdf8430636e5bf4afcccf3 | | BAN | 76 | Not recorded | 699bdf1909c33900fcd804ee | | MUMC | 75.61 | Not recorded | 69cdf8430636e5bf4afcccf3 | | Robust-MMR | 75.6 | Not recorded | 699bdf1909c33900fcd804ee | | Qwen2.5-VL-32B-Instruct | 75.12 | Not recorded | 68fa7aa61e209131cea84677 | | CCIS-MVQA\* | 75.06 | Not recorded | 69cdf8430636e5bf4afcccf3 | | MedVR | 74.4 | Not recorded | 69d884327d0172140ec77b60 | | HuatuoGPT-Vision-34B | 74.26 | Not recorded | 68fa7aa61e209131cea84677 | | ToolTree | 74.12 | Not recorded | 69b78e650636e5bf4afbda98 | | Qwen2.5-VL-7B-Instruct | 74 | Not recorded | 69a66c1a2d603e6de3e3ce72 | | QoQ-Med-7B | 74 | Not recorded | 69a66c1a2d603e6de3e3ce72 | | Qwen3vl-8B | 73.71 | Not recorded | 6965d1af6343fa78c83238cb | | MedVLSynther-3B | 73.53 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | IBISAgent | 73.4 | Not recorded | 695de86a53fdd5c361743b05 | | GPT-5-mini | 73.31 | Not recorded | 6965d1af6343fa78c83238cb | | MPR | 73.3 | Not recorded | 69cdf8430636e5bf4afcccf3 | | UnICLAM\* | 73.2 | Not recorded | 69cdf8430636e5bf4afcccf3 | | InternVL3-8B | 72.91 | Not recorded | 6965d1af6343fa78c83238cb | | Chiron | 72.7 | Not recorded | 695de86a53fdd5c361743b05 | | MedGemma 27B | 72.67 | Not recorded | 68fa7aa61e209131cea84677 | | MedGemma 27B | 72.67 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | Direct GRPO w/o cold-start | 72.64 | Not recorded | 6965d1af6343fa78c83238cb | | ViToS | 72.51 | Not recorded | 6a449f14e0224bd4f4993152 | | Chiron-01-8B | 72.5 | Not recorded | 6965d1af6343fa78c83238cb | | MedGemma-4B | 72.5 | Not recorded | 69d884327d0172140ec77b60 | | Qwen2.5-VL-32B | 71.8 | Not recorded | 69d884327d0172140ec77b60 | | Qwen2.5vl-32B | 71.7 | Not recorded | 6965d1af6343fa78c83238cb | | Qwen2.5-VL-32B | 71.7 | Not recorded | 69ddca087d0172140ec79977 | | Gemini-2.5-pro | 71.43 | Not recorded | 6965d1af6343fa78c83238cb | | Lingshu-32B | 71.4 | Not recorded | 68fa7c471e209131cea8d5f5 | | ViToS | 71.31 | Not recorded | 6a449f14e0224bd4f4993152 | | ViToS | 71.3 | Not recorded | 6a449f14e0224bd4f4993152 | | MedVLThinker-3B RL | 71.08 | Not recorded | 68fa7aa61e209131cea84677 | | MedVLThinker-3B | 71.08 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | LINGSHU-7B + MED-SCOUT | 71 | Not recorded | 69802fdda8b01e76cc8e185a | | CPRD | 71 | Not recorded | 699bdf1909c33900fcd804ee | | MEDGEMMA-4B-IT | 70.8 | Not recorded | 69802fdda8b01e76cc8e185a | | GRPO w/o Tools | 70.75 | Not recorded | 6965d1af6343fa78c83238cb | | Cold-start w/o Tools | 70.75 | Not recorded | 6965d1af6343fa78c83238cb | | MEDVISTA-R1 | 70.75 | Not recorded | 6965d1af6343fa78c83238cb | | MEDVISTA-R1 | 70.75 | Not recorded | 6965d1af6343fa78c83238cb | | ViToS | 70.52 | Not recorded | 6a449f14e0224bd4f4993152 | | Baseline | 70.52 | Not recorded | 6a449f14e0224bd4f4993152 | | VQAMix\* | 70.5 | Not recorded | 69cdf8430636e5bf4afcccf3 | | GPT-4o | 70.22 | Not recorded | 68fa7aa61e209131cea84677 | | GEMINI-3-FLASH | 70.2 | Not recorded | 69802fdda8b01e76cc8e185a | | HUATUOGPT-VISION-7B + MED-SCOUT | 70.1 | Not recorded | 69802fdda8b01e76cc8e185a | | ViTAR | 70.1 | Not recorded | 6a449f14e0224bd4f4993152 | | MEVF | 69.4 | Not recorded | 699bdf1909c33900fcd804ee | | M2I2 | 69.16 | Not recorded | 69cdf8430636e5bf4afcccf3 | | LINGSHU-7B | 68.9 | Not recorded | 69802fdda8b01e76cc8e185a | | Qwen2.5-VL-7B-Instruct | 68.75 | Not recorded | 68fa7aa61e209131cea84677 | | Qwen2.5-VL-7B-Instruct | 68.75 | Not recorded | 6904426f3ef7ccd74e5e4a72 | | CARE-Coord-B | 68.29 | Not recorded | 69a66c632d603e6de3e3d242 | | BAN | 68.2 | Not recorded | 699bdf1909c33900fcd804ee | | Lingshu | 67.9 | Not recorded | 69cb54f10636e5bf4afc9dbf | | Lingshu-7B | 67.9 | Not recorded | 69d884327d0172140ec77b60 | | Lingshu-7B | 67.9 | Not recorded | 69f190b655a3c32af89fbc0e | | VScan | 67.73 | Not recorded | 6a449f14e0224bd4f4993152 | | Rform + Rexact + RDTW | 67.6 | Not recorded | 69fc1c7d5497b3b670c88460 | | MedGemma-4B | 67.6 | Not recorded | 68fa7c471e209131cea8d5f5 | | Claude Sonnet 4 | 67.6 | Not recorded | 68fa7c471e209131cea8d5f5 | | GPT-5 | 67.53 | Not recorded | 6965d1af6343fa78c83238cb | | Original | 67.41 | Not recorded | 69608b6d53fdd5c361744271 | | Gemini-2.5-Flash | 67.41 | Not recorded | 68fa7c471e209131cea8d5f5 | | VisionZip | 67.33 | Not recorded | 6a449f14e0224bd4f4993152 | | HUATUOGPT-VISION-7B | 67 | Not recorded | 69802fdda8b01e76cc8e185a | ### Robust-MMR Accuracy: 80.4. Result ID: 699cde6b5698665ec47f229a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Standard"}. This record does not assert independent reproduction. ### OctoMed-7B Accuracy: 79. Result ID: 69a836846dbb7075b94ec33c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Ours (7B) Accuracy: 79. Result ID: 69a836846dbb7075b94ec33f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### MedGemma 4B Accuracy: 78.55. Result ID: 699ae4335698665ec474472e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### MedGemma 4B Accuracy: 78.55. Result ID: 699ae6545698665ec47695bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Medical LLM"}. This record does not assert independent reproduction. ### KG-CMI Accuracy: 78.21. Result ID: 69ce3b3847f34ce924b7d47e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### CPRD Accuracy: 78.2. Result ID: 699cde6b5698665ec47f2298. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Standard"}. This record does not assert independent reproduction. ### LaPA Accuracy: 78.15. Result ID: 69ce3b3847f34ce924b7d478. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### MedVLSynther-7B Accuracy: 77.57. Result ID: 699ae6545698665ec47695c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Medical LLM"}. This record does not assert independent reproduction. ### Ours-7B Accuracy: 77.16. Result ID: 69f1baa347f34ce924d80ac6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medical Models"}. This record does not assert independent reproduction. ### MEVF Accuracy: 77. Result ID: 699cde6b5698665ec47f2296. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Standard"}. This record does not assert independent reproduction. ### MedVLThinker-32B RL Accuracy: 76.96. Result ID: 699ae4335698665ec4744735. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### M3AE Accuracy: 76.72. Result ID: 69ce3b3847f34ce924b7d474. [Reporting paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### VG-CALF Accuracy: 76.16. Result ID: 69ce3b3847f34ce924b7d47c. [Reporting paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### BAN Accuracy: 76. Result ID: 699cde6b5698665ec47f2294. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Standard"}. This record does not assert independent reproduction. ### MUMC Accuracy: 75.61. Result ID: 69ce3b3847f34ce924b7d473. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### Robust-MMR Accuracy: 75.6. Result ID: 699cde6b5698665ec47f229b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Perturbed"}. This record does not assert independent reproduction. ### Qwen2.5-VL-32B-Instruct Accuracy: 75.12. Result ID: 699ae4335698665ec474472d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### CCIS-MVQA\* Accuracy: 75.06. Result ID: 69ce3b3847f34ce924b7d47d. [Reporting paper]() Recorded evaluation configuration: {"reported\_format":"Point estimate"}. This record does not assert independent reproduction. ### MedVR Accuracy: 74.4. Result ID: 69d8bebb47f34ce924c1d952. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medical-Specific VLMs","Evaluation Protocol":"In-domain","Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### HuatuoGPT-Vision-34B Accuracy: 74.26. Result ID: 699ae4335698665ec4744732. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### ToolTree Accuracy: 74.12. Result ID: 69b7c9fe47f34ce9249faebe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"GPT-4o"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-Instruct Accuracy: 74. Result ID: 69a836846dbb7075b94ec339. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### QoQ-Med-7B Accuracy: 74. Result ID: 69a836846dbb7075b94ec33a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Qwen3vl-8B Accuracy: 73.71. Result ID: 6997cc685698665ec423a154. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters"}. This record does not assert independent reproduction. ### MedVLSynther-3B Accuracy: 73.53. Result ID: 699ae6545698665ec47695c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Medical LLM"}. This record does not assert independent reproduction. ### IBISAgent Accuracy: 73.4. Result ID: 6997e2155698665ec43af2c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5-mini Accuracy: 73.31. Result ID: 6997cc685698665ec423a14c. [Reporting paper]() Recorded evaluation configuration: {"Model size":"Proprietary","Prompting protocol":"vanilla prompt"}. This record does not assert independent reproduction. ### MPR Accuracy: 73.3. Result ID: 69ce3b3847f34ce924b7d477. [Reporting paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### UnICLAM\* Accuracy: 73.2. Result ID: 69ce3b3847f34ce924b7d475. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reported\_format":"Point estimate"}. This record does not assert independent reproduction. ### InternVL3-8B Accuracy: 72.91. Result ID: 6997cc685698665ec423a156. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters"}. This record does not assert independent reproduction. ### Chiron Accuracy: 72.7. Result ID: 6997e2155698665ec43af2c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MedGemma 27B Accuracy: 72.67. Result ID: 699ae4335698665ec474472f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### MedGemma 27B Accuracy: 72.67. Result ID: 699ae6545698665ec47695be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Medical LLM"}. This record does not assert independent reproduction. ### Direct GRPO w/o cold-start Accuracy: 72.64. Result ID: 6997cc685698665ec423a159. [Reporting paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters","SFT-stage setting":"removed","Reasoning-stage setting":"Direct GRPO"}. This record does not assert independent reproduction. ### ViToS Accuracy: 72.51. Result ID: 6a44cca373d097d845bf31ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Lingshu-32B"}. This record does not assert independent reproduction. ### Chiron-01-8B Accuracy: 72.5. Result ID: 6997cc685698665ec423a1f6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tool":false}. This record does not assert independent reproduction. ### MedGemma-4B Accuracy: 72.5. Result ID: 69d8bebb47f34ce924c1d94e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medical-Specific VLMs","Evaluation Protocol":"In-domain"}. This record does not assert independent reproduction. ### Qwen2.5-VL-32B Accuracy: 71.8. Result ID: 69d8bebb47f34ce924c1d948. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"General-Purpose VLMs","Evaluation Protocol":"In-domain"}. This record does not assert independent reproduction. ### Qwen2.5vl-32B Accuracy: 71.7. Result ID: 6997cc685698665ec423a1f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tool":false}. This record does not assert independent reproduction. ### Qwen2.5-VL-32B Accuracy: 71.7. Result ID: 69dded2b47f34ce924c50344. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-Source SOTA"}. This record does not assert independent reproduction. ### Gemini-2.5-pro Accuracy: 71.43. Result ID: 6997cc685698665ec423a150. [Reporting paper]() Recorded evaluation configuration: {"Model size":"Proprietary","Prompting protocol":"vanilla prompt"}. This record does not assert independent reproduction. ### Lingshu-32B Accuracy: 71.4. Result ID: 6a1d39edfbc373592c16248f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"32B","Medical":"yes"}. This record does not assert independent reproduction. ### ViToS Accuracy: 71.31. Result ID: 6a44cca373d097d845bf31a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Lingshu-7B"}. This record does not assert independent reproduction. ### ViToS Accuracy: 71.3. Result ID: 6a44cca373d097d845bf31f8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MedVLThinker-3B RL Accuracy: 71.08. Result ID: 699ae4335698665ec4744733. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### MedVLThinker-3B Accuracy: 71.08. Result ID: 699ae6545698665ec47695c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Medical LLM"}. This record does not assert independent reproduction. ### LINGSHU-7B + MED-SCOUT Accuracy: 71. Result ID: 69976ed45698665ec4c2aecc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B","Enhancement":"Med-Scout"}. This record does not assert independent reproduction. ### CPRD Accuracy: 71. Result ID: 699cde6b5698665ec47f2299. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Perturbed"}. This record does not assert independent reproduction. ### MEDGEMMA-4B-IT Accuracy: 70.8. Result ID: 69976ed45698665ec4c2aec8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction. ### GRPO w/o Tools Accuracy: 70.75. Result ID: 6997cc685698665ec423a158. [Reporting paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters","Tool-access setting":"removed","Reasoning-stage setting":"GRPO"}. This record does not assert independent reproduction. ### Cold-start w/o Tools Accuracy: 70.75. Result ID: 6997cc685698665ec423a15a. [Reporting paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters","Tool-access setting":"removed","SFT-stage setting":"Cold-start"}. This record does not assert independent reproduction. ### MEDVISTA-R1 Accuracy: 70.75. Result ID: 6997cc685698665ec423a15c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"7-13B parameters"}. This record does not assert independent reproduction. ### MEDVISTA-R1 Accuracy: 70.75. Result ID: 6997cc685698665ec423a1fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tool":true}. This record does not assert independent reproduction. ### ViToS Accuracy: 70.52. Result ID: 6a44cca373d097d845bf31a7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"HuatuoGPT-Vision-7B"}. This record does not assert independent reproduction. ### Baseline Accuracy: 70.52. Result ID: 6a44cca373d097d845bf31a8. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Lingshu-32B"}. This record does not assert independent reproduction. ### VQAMix\* Accuracy: 70.5. Result ID: 69ce3b3847f34ce924b7d47b. [Reporting paper]() Recorded evaluation configuration: {"reported\_format":"Point estimate"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 70.22. Result ID: 699ae4335698665ec4744728. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### GEMINI-3-FLASH Accuracy: 70.2. Result ID: 69976ed45698665ec4c2aebf. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Proprietary"}. This record does not assert independent reproduction. ### HUATUOGPT-VISION-7B + MED-SCOUT Accuracy: 70.1. Result ID: 69976ed45698665ec4c2aeca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B","Enhancement":"Med-Scout"}. This record does not assert independent reproduction. ### ViTAR Accuracy: 70.1. Result ID: 6a44cca373d097d845bf31f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MEVF Accuracy: 69.4. Result ID: 699cde6b5698665ec47f2297. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Perturbed"}. This record does not assert independent reproduction. ### M2I2 Accuracy: 69.16. Result ID: 69ce3b3847f34ce924b7d472. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reported\_format":"Mean ± Standard Deviation"}. This record does not assert independent reproduction. ### LINGSHU-7B Accuracy: 68.9. Result ID: 69976ed45698665ec4c2aecb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-Instruct Accuracy: 68.75. Result ID: 699ae4335698665ec474472c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding":"Greedy"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-Instruct Accuracy: 68.75. Result ID: 699ae6545698665ec47695bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"General LLM"}. This record does not assert independent reproduction. ### CARE-Coord-B Accuracy: 68.29. Result ID: 69a842166dbb7075b94f1433. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Type":"Open-source"}. This record does not assert independent reproduction. ### BAN Accuracy: 68.2. Result ID: 699cde6b5698665ec47f2295. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Perturbed"}. This record does not assert independent reproduction. ### Lingshu Accuracy: 67.9. Result ID: 69cb83ae47f34ce924b3bca7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### Lingshu-7B Accuracy: 67.9. Result ID: 69d8bebb47f34ce924c1d951. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medical-Specific VLMs","Evaluation Protocol":"In-domain"}. This record does not assert independent reproduction. ### Lingshu-7B Accuracy: 67.9. Result ID: 69f1baa347f34ce924d80ac5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medical Models"}. This record does not assert independent reproduction. ### VScan Accuracy: 67.73. Result ID: 6a44cca373d097d845bf31a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"HuatuoGPT-Vision-7B"}. This record does not assert independent reproduction. ### Rform + Rexact + RDTW Accuracy: 67.6. Result ID: 69fc3dd347f34ce924de929a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reward functions":"Rform + Rexact + RDTW"}. This record does not assert independent reproduction. ### MedGemma-4B Accuracy: 67.6. Result ID: 6a1d39edfbc373592c16248d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"4B","Medical":"yes"}. This record does not assert independent reproduction. ### Claude Sonnet 4 Accuracy: 67.6. Result ID: 6a1d39edfbc373592c162492. [Reporting paper]() Recorded evaluation configuration: {"Params":"Closed","Medical":"no"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 67.53. Result ID: 6997cc685698665ec423a14b. [Reporting paper]() Recorded evaluation configuration: {"Model size":"Proprietary","Prompting protocol":"vanilla prompt"}. This record does not assert independent reproduction. ### Original Accuracy: 67.41. Result ID: 699985775698665ec4f65103. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Lingshu-7B"}. This record does not assert independent reproduction. ### Gemini-2.5-Flash Accuracy: 67.41. Result ID: 6a1d39edfbc373592c162493. [Reporting paper]() Recorded evaluation configuration: {"Params":"Closed","Medical":"no"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 67.33. Result ID: 6a44cca373d097d845bf31a3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"HuatuoGPT-Vision-7B"}. This record does not assert independent reproduction. ### HUATUOGPT-VISION-7B Accuracy: 67. Result ID: 69976ed45698665ec4c2aec9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. Page 1 of 4. This is an index page, not the full collection. - [Next index page](): Follow this link to discover more pages.