# Multimodal Reranking on MMEB VisDoc v2 Source: [Multimodal Reranking on MMEB VisDoc v2]() Collection index: [llms.txt]() Data updated: 2026-06-11T08:35:41.584Z Evaluation of Multimodal Reranking performance using metrics VDRv1, VDRv2, VR, OOD, and Overall on the MMEB VisDoc v2 dataset. 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 | Multimodal Reranking | | Dataset (including any stated split) | MMEB VisDoc v2 | | Metric | Overall Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 Score](): Higher is better; unit: PERCENT. - [VDRv1](): Higher is better; unit: PERCENT. - [VDRv2](): Higher is better; unit: PERCENT. - [VR](): Higher is better; unit: PERCENT. - [OOD](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen3-VL-8B-reranker | 87.1 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-8B-miniReranker | 86.7 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-4B-reranker | 86.5 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-4B-miniReranker | 86.2 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-Reranker-8B | 85.8 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-2B-reranker | 85.7 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-2B-miniReranker | 84.9 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-Reranker-2B | 83.2 | Not recorded | 6a28f37890966ade919c19ba | | Qwen3-VL-2B-Instruct | 81.2 | Not recorded | 6a28f37890966ade919c19ba | | GME (7B) | 75.2 | Not recorded | 6a28f37890966ade919c19ba | | GME (2B) | 72.7 | Not recorded | 6a28f37890966ade919c19ba | | ColPali v1.3 (3B) | 71 | Not recorded | 6a28f37890966ade919c19ba | | LamRA-Qwen2.5 (7B) | 50.2 | Not recorded | 6a28f37890966ade919c19ba | | VLM2Vec-Qwen2VL (7B) | 46.4 | Not recorded | 6a28f37890966ade919c19ba | | VLM2Vec (2B) | 41.6 | Not recorded | 6a28f37890966ade919c19ba | ### Qwen3-VL-8B-reranker Overall Score: 87.1. Result ID: 6a2a7275fbc373592c278ffc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"8B","Fine-tuning Recipe":"our training recipe"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-miniReranker Overall Score: 86.7. Result ID: 6a2a7275fbc373592c278ffd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"8B","Fine-tuning Recipe":"our training recipe","Reranker Type":"miniReranker"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-reranker Overall Score: 86.5. Result ID: 6a2a7275fbc373592c278ff6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"4B","Fine-tuning Recipe":"our training recipe"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-miniReranker Overall Score: 86.2. Result ID: 6a2a7275fbc373592c278ff7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"4B","Fine-tuning Recipe":"our training recipe","Reranker Type":"miniReranker"}. This record does not assert independent reproduction. ### Qwen3-VL-Reranker-8B Overall Score: 85.8. Result ID: 6a2a7275fbc373592c278ffb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"8B"}. This record does not assert independent reproduction. ### Qwen3-VL-2B-reranker Overall Score: 85.7. Result ID: 6a2a7275fbc373592c278ff4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"2B","Fine-tuning Recipe":"our training recipe"}. This record does not assert independent reproduction. ### Qwen3-VL-2B-miniReranker Overall Score: 84.9. Result ID: 6a2a7275fbc373592c278ff5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"2B","Fine-tuning Recipe":"our training recipe","Reranker Type":"miniReranker"}. This record does not assert independent reproduction. ### Qwen3-VL-Reranker-2B Overall Score: 83.2. Result ID: 6a2a7275fbc373592c278ff2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"2B"}. This record does not assert independent reproduction. ### Qwen3-VL-2B-Instruct Overall Score: 81.2. Result ID: 6a2a7275fbc373592c278ff3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"2B"}. This record does not assert independent reproduction. ### GME (7B) Overall Score: 75.2. Result ID: 6a2a7275fbc373592c278ff8. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"7B"}. This record does not assert independent reproduction. ### GME (2B) Overall Score: 72.7. Result ID: 6a2a7275fbc373592c278ff0. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"2B"}. This record does not assert independent reproduction. ### ColPali v1.3 (3B) Overall Score: 71. Result ID: 6a2a7275fbc373592c278fef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B"}. This record does not assert independent reproduction. ### LamRA-Qwen2.5 (7B) Overall Score: 50.2. Result ID: 6a2a7275fbc373592c278ff9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"7B"}. This record does not assert independent reproduction. ### VLM2Vec-Qwen2VL (7B) Overall Score: 46.4. Result ID: 6a2a7275fbc373592c278ffa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B"}. This record does not assert independent reproduction. ### VLM2Vec (2B) Overall Score: 41.6. Result ID: 6a2a7275fbc373592c278ff1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"2B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.