# Referring Expression Comprehension on RefCOCO (F1 Average)

Source: [Referring Expression Comprehension on RefCOCO (F1 Average)](<https://www.sota2.com/research/sota/referring-expression-comprehension-on-refcoco-f1-average>)

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

Data updated: 2026-05-26T11:59:27.681Z

Evaluation of Referring Expression Comprehension performance on the RefCOCO dataset, specifically reporting the F1 Average metric.

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 | Referring Expression Comprehension |
| Dataset (including any stated split) | RefCOCO |
| Metric | F1 (RefCOCO) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [F1 (RefCOCO)](<https://www.sota2.com/research/sota/referring-expression-comprehension-on-refcoco-f1-average.md?metric=F1%20(RefCOCO)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 (RefCOCO) | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen3-VL | 94.1 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Qwen2.5-VL | 93.5 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Qwen3-VL | 92.4 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Qwen2.5-VL | 91.2 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Qwen3-VL | 90.8 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| GLM-4.6V | 90.1 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Qwen2-VL | 88.4 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| MiniCPM-V | 87.2 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| LLaVA-OV-1.5 | 86.9 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| InternVL3.5 | 85.6 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Grounding-Dino | 85.1 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Keye-VL-1.5 | 84.5 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| MiMo-VL-RL | 83 | Not recorded | 6a15294bdbd7150506f7e1d6 |
| Mistral-3 | 32.1 | Not recorded | 6a15294bdbd7150506f7e1d6 |

### Qwen3-VL

F1 (RefCOCO): 94.1. Result ID: 6a158b18fbc373592c0e1f54.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

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

### Qwen2.5-VL

F1 (RefCOCO): 93.5. Result ID: 6a158b18fbc373592c0e1f4a.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Size":"32B"}. This record does not assert independent reproduction.

### Qwen3-VL

F1 (RefCOCO): 92.4. Result ID: 6a158b18fbc373592c0e1f53.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction.

### Qwen2.5-VL

F1 (RefCOCO): 91.2. Result ID: 6a158b18fbc373592c0e1f49.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

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

### Qwen3-VL

F1 (RefCOCO): 90.8. Result ID: 6a158b18fbc373592c0e1f52.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

Recorded evaluation configuration: {"Size":"2B"}. This record does not assert independent reproduction.

### GLM-4.6V

F1 (RefCOCO): 90.1. Result ID: 6a158b18fbc373592c0e1f4c.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aaf1e209131cea849ba.md>)

Recorded evaluation configuration: {"Size":"9B"}. This record does not assert independent reproduction.

### Qwen2-VL

F1 (RefCOCO): 88.4. Result ID: 6a158b18fbc373592c0e1f48.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

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

### MiniCPM-V

F1 (RefCOCO): 87.2. Result ID: 6a158b18fbc373592c0e1f4e.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa4baa1e209131ce94799b.md>)

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

### LLaVA-OV-1.5

F1 (RefCOCO): 86.9. Result ID: 6a158b18fbc373592c0e1f4b.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c031e209131cea8bdde.md>)

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

### InternVL3.5

F1 (RefCOCO): 85.6. Result ID: 6a158b18fbc373592c0e1f4f.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7af01e209131cea85dc9.md>)

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

### Grounding-Dino

F1 (RefCOCO): 85.1. Result ID: 6a158b18fbc373592c0e1f47.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74311e209131cea6055d.md>)

[Implementation](<https://github.com/IDEA-Research/GroundingDINO>)

Recorded evaluation configuration: {"Size":"—"}. This record does not assert independent reproduction.

### Keye-VL-1.5

F1 (RefCOCO): 84.5. Result ID: 6a158b18fbc373592c0e1f4d.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b2d1e209131cea87107.md>)

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

### MiMo-VL-RL

F1 (RefCOCO): 83. Result ID: 6a158b18fbc373592c0e1f51.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79121e209131cea7bfbd.md>)

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

### Mistral-3

F1 (RefCOCO): 32.1. Result ID: 6a158b18fbc373592c0e1f50.

[Reporting paper](<https://www.sota2.com/research/paper/6a15294bdbd7150506f7e1d6.md>)

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

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