# Hallucination object detection on HOD

Source: [Hallucination object detection on HOD](<https://www.sota2.com/research/sota/hallucination-object-detection-on-hod>)

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

Data updated: 2026-06-25T07:48:11.317Z

Evaluation of model performance on hallucination object detection using the HOD dataset, measured by 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 | Hallucination object detection |
| Dataset (including any stated split) | HOD |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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/hallucination-object-detection-on-hod.md?metric=Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| Yuvion VL | 98.9 | Not recorded | 6a3cb602206542251d7ca4df |
| Yuvion VL | 95.2 | Not recorded | 6a3cb602206542251d7ca4df |
| Yuvion VL | 92.8 | Not recorded | 6a3cb602206542251d7ca4df |
| Yuvion VL | 79.2 | Not recorded | 6a3cb602206542251d7ca4df |
| K2.5 | 56.7 | Not recorded | 6a3cb602206542251d7ca4df |
| Qwen3-VL | 52.5 | Not recorded | 6a3cb602206542251d7ca4df |
| Qwen3-VL | 46.8 | Not recorded | 6a3cb602206542251d7ca4df |
| Qwen3-VL | 33.2 | Not recorded | 6a3cb602206542251d7ca4df |
| GPT-5.4 | 32.4 | Not recorded | 6a3cb602206542251d7ca4df |
| Qwen3.5-Plus | 32.1 | Not recorded | 6a3cb602206542251d7ca4df |
| Qwen3-VL | 31.8 | Not recorded | 6a3cb602206542251d7ca4df |
| Opus-4.6 | 14.4 | Not recorded | 6a3cb602206542251d7ca4df |

### Yuvion VL

Score: 98.9. Result ID: 6a3cdd29fbc373592c51ae94.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"32B","Decoding Mode":"Instruct"}. This record does not assert independent reproduction.

### Yuvion VL

Score: 95.2. Result ID: 6a3cdd29fbc373592c51ae91.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"8B","Decoding Mode":"Reasoning"}. This record does not assert independent reproduction.

### Yuvion VL

Score: 92.8. Result ID: 6a3cdd29fbc373592c51ae90.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"8B","Decoding Mode":"Instruct"}. This record does not assert independent reproduction.

### Yuvion VL

Score: 79.2. Result ID: 6a3cdd29fbc373592c51ae95.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"32B","Decoding Mode":"Reasoning"}. This record does not assert independent reproduction.

### K2.5

Score: 56.7. Result ID: 6a3cdd29fbc373592c51ae98.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"≈1T"}. This record does not assert independent reproduction.

### Qwen3-VL

Score: 52.5. Result ID: 6a3cdd29fbc373592c51ae8f.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Scale":"8B","Decoding Mode":"Thinking"}. This record does not assert independent reproduction.

### Qwen3-VL

Score: 46.8. Result ID: 6a3cdd29fbc373592c51ae92.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Scale":"32B","Decoding Mode":"Instruct"}. This record does not assert independent reproduction.

### Qwen3-VL

Score: 33.2. Result ID: 6a3cdd29fbc373592c51ae93.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Scale":"32B","Decoding Mode":"Thinking"}. This record does not assert independent reproduction.

### GPT-5.4

Score: 32.4. Result ID: 6a3cdd29fbc373592c51ae96.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"≈1.5T"}. This record does not assert independent reproduction.

### Qwen3.5-Plus

Score: 32.1. Result ID: 6a3cdd29fbc373592c51ae99.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"397B"}. This record does not assert independent reproduction.

### Qwen3-VL

Score: 31.8. Result ID: 6a3cdd29fbc373592c51ae8e.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Scale":"8B","Decoding Mode":"Instruct"}. This record does not assert independent reproduction.

### Opus-4.6

Score: 14.4. Result ID: 6a3cdd29fbc373592c51ae97.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb602206542251d7ca4df.md>)

Recorded evaluation configuration: {"Model Scale":"≈800B"}. This record does not assert independent reproduction.

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