# Real-world cross-view spatial reasoning on MMSI Source: [Real-world cross-view spatial reasoning on MMSI]() Collection index: [llms.txt]() Data updated: 2026-05-27T10:21:12.995Z Evaluation of model performance on the MMSI dataset focusing on Real-world cross-view spatial reasoning capabilities. 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 | Real-world cross-view spatial reasoning | | Dataset (including any stated split) | MMSI | | Metric | Overall Accuracy | | 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 - [Overall Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Top-down + VDrop | 32 | Not recorded | 6a167ad9dbd7150506f802fe | | Top-down | 28.8 | Not recorded | 6a167ad9dbd7150506f802fe | | Point Matching + VDrop | 28.3 | Not recorded | 6a167ad9dbd7150506f802fe | | Qwen3-VL-8B | 28 | Not recorded | 6a167ad9dbd7150506f802fe | | Point Matching | 27.8 | Not recorded | 6a167ad9dbd7150506f802fe | | Qwen3-VL-4B | 27.4 | Not recorded | 6a167ad9dbd7150506f802fe | | No-Think | 27.4 | Not recorded | 6a167ad9dbd7150506f802fe | | BAGEL | 26.9 | Not recorded | 6a167ad9dbd7150506f802fe | | ThinkMorph (24K) | 26.5 | Not recorded | 6a167ad9dbd7150506f802fe | | Panoramic + VDrop | 26 | Not recorded | 6a167ad9dbd7150506f802fe | | Panoramic | 24.9 | Not recorded | 6a167ad9dbd7150506f802fe | | Text CoT | 24.6 | Not recorded | 6a167ad9dbd7150506f802fe | | BAGEL-Zebra-CoT (182K) | 23.2 | Not recorded | 6a167ad9dbd7150506f802fe | ### Top-down + VDrop Overall Accuracy: 32. Result ID: 6a16c560fbc373592c108b51. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### Top-down Overall Accuracy: 28.8. Result ID: 6a16c560fbc373592c108b50. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Point Matching + VDrop Overall Accuracy: 28.3. Result ID: 6a16c560fbc373592c108b4f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Overall Accuracy: 28. Result ID: 6a16c560fbc373592c108b46. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Point Matching Overall Accuracy: 27.8. Result ID: 6a16c560fbc373592c108b4e. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Qwen3-VL-4B Overall Accuracy: 27.4. Result ID: 6a16c560fbc373592c108b45. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### No-Think Overall Accuracy: 27.4. Result ID: 6a16c560fbc373592c108b4a. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### BAGEL Overall Accuracy: 26.9. Result ID: 6a16c560fbc373592c108b47. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### ThinkMorph (24K) Overall Accuracy: 26.5. Result ID: 6a16c560fbc373592c108b49. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Panoramic + VDrop Overall Accuracy: 26. Result ID: 6a16c560fbc373592c108b4d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### Panoramic Overall Accuracy: 24.9. Result ID: 6a16c560fbc373592c108b4c. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Text CoT Overall Accuracy: 24.6. Result ID: 6a16c560fbc373592c108b4b. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### BAGEL-Zebra-CoT (182K) Overall Accuracy: 23.2. Result ID: 6a16c560fbc373592c108b48. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.