# Visual Question Answering on HallusionBench HBI (all)

Source: [Visual Question Answering on HallusionBench HBI (all)](<https://www.sota2.com/research/sota/visual-question-answering-on-hallusionbench-hbi-all>)

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

Data updated: 2026-02-26T06:03:46.808Z

Evaluation of model performance on the HallusionBench HBI (all) dataset for Visual Question Answering.

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 Question Answering |
| Dataset (including any stated split) | HallusionBench HBI (all) |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Score](<https://www.sota2.com/research/sota/visual-question-answering-on-hallusionbench-hbi-all.md?metric=Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| MLCD | 45.21 | Not recorded | 68fa758d1e209131cea688d2 |
| CLIP | 45.1 | Not recorded | 68fa758d1e209131cea688d2 |
| CLIP | 39.96 | Not recorded | 68fa758d1e209131cea688d2 |
| MLCD | 39.46 | Not recorded | 68fa758d1e209131cea688d2 |

### MLCD

Score: 45.21. Result ID: 6993d2683a6c31484918c560.

[Reporting paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Implementation](<https://github.com/deepglint/unicom>)

Recorded evaluation configuration: {"LLM":"Qwen2-72B","Vision Tower":"ViT-L/14"}. This record does not assert independent reproduction.

### CLIP

Score: 45.1. Result ID: 6993d2683a6c31484918c55f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee41e209131cea42adf.md>)

[Implementation](<https://github.com/facebookresearch/flip>)

Recorded evaluation configuration: {"LLM":"Qwen2-72B","Vision Tower":"ViT-L/14"}. This record does not assert independent reproduction.

### CLIP

Score: 39.96. Result ID: 6993d2683a6c31484918c55d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee41e209131cea42adf.md>)

[Implementation](<https://github.com/facebookresearch/flip>)

Recorded evaluation configuration: {"LLM":"Qwen2-7B","Vision Tower":"ViT-L/14"}. This record does not assert independent reproduction.

### MLCD

Score: 39.46. Result ID: 6993d2683a6c31484918c55e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758d1e209131cea688d2.md>)

[Implementation](<https://github.com/deepglint/unicom>)

Recorded evaluation configuration: {"LLM":"Qwen2-7B","Vision Tower":"ViT-L/14"}. This record does not assert independent reproduction.

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