# Visual Reasoning on DeepTumorVQA Source: [Visual Reasoning on DeepTumorVQA]() Collection index: [llms.txt]() Data updated: 2026-03-27T12:51:52.113Z Evaluation of model performance on visual reasoning capabilities using the DeepTumorVQA dataset, focusing on lesion characteristics, organ comparisons, and spatial relationships. 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 | Visual Reasoning | | Dataset (including any stated split) | DeepTumorVQA | | Metric | Adjacent Organ Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 - [Adjacent Organ Score](): Higher is better; unit: PERCENT. - [Inter-Segment Comparison Score](): Higher is better; unit: PERCENT. - [Kidney Volume Comparison Score](): Higher is better; unit: PERCENT. - [Largest Lesion Attenuation Score](): Higher is better; unit: PERCENT. - [Largest Lesion Diameter Score](): Higher is better; unit: PERCENT. - [Largest Lesion Location Score](): Higher is better; unit: PERCENT. - [Largest Lesion Slice Score](): Higher is better; unit: PERCENT. - [Lesion Count by Location Score](): Higher is better; unit: PERCENT. - [Lesion Counting Score](): Higher is better; unit: PERCENT. - [Lesion Outlier Score](): Higher is better; unit: PERCENT. - [Liver Lesion Clustering Score](): Higher is better; unit: PERCENT. - [Organ Aggregation Score](): Higher is better; unit: PERCENT. - [Organ Enlargement Score](): Higher is better; unit: PERCENT. - [Tumor Organ HU Difference Score](): Higher is better; unit: PERCENT. - [Average Performance Score](): Higher is better; unit: PERCENT. - [Accuracy](): Higher is better; unit: PERCENT. - [MRA Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Adjacent Organ Score | Result date | Report paper | | --- | --- | --- | --- | | Photon | 69.6 | Not recorded | 69c60efd0636e5bf4afc84c6 | | M3D-P3 | 60.9 | Not recorded | 69c60efd0636e5bf4afc84c6 | | CT-CHAT | 60.9 | Not recorded | 69c60efd0636e5bf4afc84c6 | | RadFM | 60.9 | Not recorded | 69c60efd0636e5bf4afc84c6 | | M3D-L2 | 56.5 | Not recorded | 69c60efd0636e5bf4afc84c6 | | RadFM | 43.5 | Not recorded | 69c60efd0636e5bf4afc84c6 | | Rand | 33.3 | Not recorded | 69c60efd0636e5bf4afc84c6 | | M3D-P3 | 30.4 | Not recorded | 69c60efd0636e5bf4afc84c6 | | CT-CHAT | 30.4 | Not recorded | 69c60efd0636e5bf4afc84c6 | | Photon | 30.4 | Not recorded | 69c60efd0636e5bf4afc84c6 | | Merlin | 21.7 | Not recorded | 69c60efd0636e5bf4afc84c6 | | Merlin | 17.4 | Not recorded | 69c60efd0636e5bf4afc84c6 | | M3D-L2 | 17.4 | Not recorded | 69c60efd0636e5bf4afc84c6 | ### Photon Adjacent Organ Score: 69.6. Result ID: 69c67d5847f34ce924b06d70. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"3B"}. This record does not assert independent reproduction. ### M3D-P3 Adjacent Organ Score: 60.9. Result ID: 69c67d5847f34ce924b06d6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"4B"}. This record does not assert independent reproduction. ### CT-CHAT Adjacent Organ Score: 60.9. Result ID: 69c67d5847f34ce924b06d6c. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"7B"}. This record does not assert independent reproduction. ### RadFM Adjacent Organ Score: 60.9. Result ID: 69c67d5847f34ce924b06d6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"13B"}. This record does not assert independent reproduction. ### M3D-L2 Adjacent Organ Score: 56.5. Result ID: 69c67d5847f34ce924b06d68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"7B"}. This record does not assert independent reproduction. ### RadFM Adjacent Organ Score: 43.5. Result ID: 69c67d5847f34ce924b06d6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"13B"}. This record does not assert independent reproduction. ### Rand Adjacent Organ Score: 33.3. Result ID: 69c67d5847f34ce924b06d65. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice"}. This record does not assert independent reproduction. ### M3D-P3 Adjacent Organ Score: 30.4. Result ID: 69c67d5847f34ce924b06d6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"4B"}. This record does not assert independent reproduction. ### CT-CHAT Adjacent Organ Score: 30.4. Result ID: 69c67d5847f34ce924b06d6d. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"7B"}. This record does not assert independent reproduction. ### Photon Adjacent Organ Score: 30.4. Result ID: 69c67d5847f34ce924b06d71. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"3B"}. This record does not assert independent reproduction. ### Merlin Adjacent Organ Score: 21.7. Result ID: 69c67d5847f34ce924b06d66. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Mode":"Multi-choice","Model Parameters":"7B"}. This record does not assert independent reproduction. ### Merlin Adjacent Organ Score: 17.4. Result ID: 69c67d5847f34ce924b06d67. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"7B"}. This record does not assert independent reproduction. ### M3D-L2 Adjacent Organ Score: 17.4. Result ID: 69c67d5847f34ce924b06d69. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Mode":"Free-text","Model Parameters":"7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.