# Multiple Choice Visual Question Answering on NegBench COCO (test)

Source: [Multiple Choice Visual Question Answering on NegBench COCO (test)](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-13T08:25:06.047Z

Evaluation of model performance on the test split of the NegBench COCO dataset for Multiple Choice Visual Question Answering capabilities, reporting metrics such as Acc, Acc Rank, CR, CR Rank, and AccE.

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 | Multiple Choice Visual Question Answering |
| Dataset (including any stated split) | NegBench COCO (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy Rank](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test.md?metric=Accuracy%20Rank>): Lower is better; unit: SCALAR.

- [CR](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test.md?metric=CR>): Higher is better; unit: PERCENT.

- [CR Rank](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test.md?metric=CR%20Rank>): Lower is better; unit: SCALAR.

- [AccE Score](<https://www.sota2.com/research/sota/multiple-choice-visual-question-answering-on-negbench-coco-test.md?metric=AccE%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| InternVL3.0-8B | 93.19 | Not recorded | 69b399f90636e5bf4afbc9ee |
| InternVL2.5-8B | 91.21 | Not recorded | 69b399f90636e5bf4afbc9ee |
| InternVL3.5-8B | 84.29 | Not recorded | 69b399f90636e5bf4afbc9ee |
| Qwen2.5-VL-7B | 81.65 | Not recorded | 69b399f90636e5bf4afbc9ee |
| Qwen3.0-VL-8B | 77.53 | Not recorded | 69b399f90636e5bf4afbc9ee |
| Gemma-3-12B | 71.98 | Not recorded | 69b399f90636e5bf4afbc9ee |
| Qwen2.0-VL-7B | 69.68 | Not recorded | 69b399f90636e5bf4afbc9ee |
| LLaVA-1.5-13B | 62.06 | Not recorded | 69b399f90636e5bf4afbc9ee |
| InternVL2.0-8B | 48.78 | Not recorded | 69b399f90636e5bf4afbc9ee |
| LLaVA-1.6-13B | 38.98 | Not recorded | 69b399f90636e5bf4afbc9ee |
| QwenVL-7B-Chat | 24.26 | Not recorded | 69b399f90636e5bf4afbc9ee |

### InternVL3.0-8B

Accuracy: 93.19. Result ID: 69b3c9d247f34ce9249dcd1c.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### InternVL2.5-8B

Accuracy: 91.21. Result ID: 69b3c9d247f34ce9249dcd1b.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### InternVL3.5-8B

Accuracy: 84.29. Result ID: 69b3c9d247f34ce9249dcd1d.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B

Accuracy: 81.65. Result ID: 69b3c9d247f34ce9249dcd22.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

[Implementation](<https://github.com/QwenLM/Qwen-VL>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

### Qwen3.0-VL-8B

Accuracy: 77.53. Result ID: 69b3c9d247f34ce9249dcd23.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

[Implementation](<https://github.com/QwenLM/Qwen-VL>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### Gemma-3-12B

Accuracy: 71.98. Result ID: 69b3c9d247f34ce9249dcd19.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

Recorded evaluation configuration: {"Parameters":"12B"}. This record does not assert independent reproduction.

### Qwen2.0-VL-7B

Accuracy: 69.68. Result ID: 69b3c9d247f34ce9249dcd21.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

[Implementation](<https://github.com/QwenLM/Qwen-VL>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

### LLaVA-1.5-13B

Accuracy: 62.06. Result ID: 69b3c9d247f34ce9249dcd1e.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Parameters":"13B"}. This record does not assert independent reproduction.

### InternVL2.0-8B

Accuracy: 48.78. Result ID: 69b3c9d247f34ce9249dcd1a.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### LLaVA-1.6-13B

Accuracy: 38.98. Result ID: 69b3c9d247f34ce9249dcd1f.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Parameters":"13B"}. This record does not assert independent reproduction.

### QwenVL-7B-Chat

Accuracy: 24.26. Result ID: 69b3c9d247f34ce9249dcd20.

[Reporting paper](<https://www.sota2.com/research/paper/69b399f90636e5bf4afbc9ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

[Implementation](<https://github.com/QwenLM/Qwen-VL>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.