# Referring Expression Grounding on RefCOCO (val)

Source: [Referring Expression Grounding on RefCOCO (val)](<https://www.sota2.com/research/sota/referring-expression-grounding-on-refcoco-val>)

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

Data updated: 2026-06-09T10:44:20.613Z

Evaluation of Referring Expression Grounding performance on the validation split of the RefCOCO dataset, reporting Accuracy at IoU threshold 0.5.

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 grounding |
| Dataset (including any stated split) | RefCOCO (val) |
| Metric | Accuracy @ IoU=0.5 |
| 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

- [Accuracy @ IoU=0.5](<https://www.sota2.com/research/sota/referring-expression-grounding-on-refcoco-val.md?metric=Accuracy%20%40%20IoU%3D0.5>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy @ IoU=0.5 | Result date | Report paper |
| --- | --- | --- | --- |
| Visual Para-Thinker++ (3B) | 90.5 | Not recorded | 6a279e5b90966ade919bfa4c |
| Visual Para-Thinker (3B) | 88.1 | Not recorded | 6a279e5b90966ade919bfa4c |
| Sequential | 85.7 | Not recorded | 6a279e5b90966ade919bfa4c |
| Qwen2.5-VL-3B | 85.4 | Not recorded | 6a279e5b90966ade919bfa4c |

### Visual Para-Thinker++ (3B)

Accuracy @ IoU=0.5: 90.5. Result ID: 6a27ed5cfbc373592c257225.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e5b90966ade919bfa4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a279e5b90966ade919bfa4c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction.

### Visual Para-Thinker (3B)

Accuracy @ IoU=0.5: 88.1. Result ID: 6a27ed5cfbc373592c257224.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e5b90966ade919bfa4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b1b1e209131cea86980.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction.

### Sequential

Accuracy @ IoU=0.5: 85.7. Result ID: 6a27ed5cfbc373592c257223.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e5b90966ade919bfa4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b1b1e209131cea86980.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction.

### Qwen2.5-VL-3B

Accuracy @ IoU=0.5: 85.4. Result ID: 6a27ed5cfbc373592c257222.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e5b90966ade919bfa4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b1b1e209131cea86980.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction.

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