# Integrating evidence across multiple images on BLINK MultiView Source: [Integrating evidence across multiple images on BLINK MultiView]() Collection index: [llms.txt]() Data updated: 2026-05-27T10:21:13.109Z Evaluation of performance on integrating evidence across multiple images using the BLINK MultiView dataset, measured by the MultiView score. 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 | Integrating evidence across multiple images | | Dataset (including any stated split) | BLINK MultiView | | Metric | MultiView Score | | Unit | SCALAR | | 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 - [MultiView Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | MultiView Score | Result date | Report paper | | --- | --- | --- | --- | | Panoramic + VDrop | 62.4 | Not recorded | 6a167ad9dbd7150506f802fe | | Top-down | 58.7 | Not recorded | 6a167ad9dbd7150506f802fe | | Top-down + VDrop | 57.1 | Not recorded | 6a167ad9dbd7150506f802fe | | Qwen3-VL-8B | 55.6 | Not recorded | 6a167ad9dbd7150506f802fe | | Panoramic | 55.6 | Not recorded | 6a167ad9dbd7150506f802fe | | Text CoT | 54.9 | Not recorded | 6a167ad9dbd7150506f802fe | | ThinkMorph (24K) | 52.6 | Not recorded | 6a167ad9dbd7150506f802fe | | Point Matching | 52.6 | Not recorded | 6a167ad9dbd7150506f802fe | | Qwen3-VL-4B | 48.9 | Not recorded | 6a167ad9dbd7150506f802fe | | BAGEL | 45.1 | Not recorded | 6a167ad9dbd7150506f802fe | | No-Think | 45.1 | Not recorded | 6a167ad9dbd7150506f802fe | | Point Matching + VDrop | 45.1 | Not recorded | 6a167ad9dbd7150506f802fe | | BAGEL-Zebra-CoT (182K) | 24.8 | Not recorded | 6a167ad9dbd7150506f802fe | ### Panoramic + VDrop MultiView Score: 62.4. Result ID: 6a16c560fbc373592c108b81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### Top-down MultiView Score: 58.7. Result ID: 6a16c560fbc373592c108b84. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Top-down + VDrop MultiView Score: 57.1. Result ID: 6a16c560fbc373592c108b85. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### Qwen3-VL-8B MultiView Score: 55.6. Result ID: 6a16c560fbc373592c108b7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Panoramic MultiView Score: 55.6. Result ID: 6a16c560fbc373592c108b80. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Text CoT MultiView Score: 54.9. Result ID: 6a16c560fbc373592c108b7f. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### ThinkMorph (24K) MultiView Score: 52.6. Result ID: 6a16c560fbc373592c108b7d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Point Matching MultiView Score: 52.6. Result ID: 6a16c560fbc373592c108b82. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Qwen3-VL-4B MultiView Score: 48.9. Result ID: 6a16c560fbc373592c108b79. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### BAGEL MultiView Score: 45.1. Result ID: 6a16c560fbc373592c108b7b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### No-Think MultiView Score: 45.1. Result ID: 6a16c560fbc373592c108b7e. [Reporting paper]() Recorded evaluation configuration: {"VDrop":"✗"}. This record does not assert independent reproduction. ### Point Matching + VDrop MultiView Score: 45.1. Result ID: 6a16c560fbc373592c108b83. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VDrop":"✓"}. This record does not assert independent reproduction. ### BAGEL-Zebra-CoT (182K) MultiView Score: 24.8. Result ID: 6a16c560fbc373592c108b7c. [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.