# Visual Grounding on RefCOCO Series

Source: [Visual Grounding on RefCOCO Series](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series>)

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

Data updated: 2026-02-26T06:01:47.215Z

Evaluation of Visual Grounding performance across the RefCOCO, RefCOCO+, and RefCOCOg datasets and their respective test splits.

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 Grounding |
| Dataset (including any stated split) | RefCOCO series |
| Metric | RefCOCO Test Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [RefCOCO Test Score](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=RefCOCO%20Test%20Score>): Higher is better; unit: PERCENT.

- [RefCOCO+ TestA Score](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=RefCOCO%2B%20TestA%20Score>): Higher is better; unit: PERCENT.

- [RefCOCO TestB Score](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=RefCOCO%20TestB%20Score>): Higher is better; unit: PERCENT.

- [RefCOCOg Test Score](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=RefCOCOg%20Test%20Score>): Higher is better; unit: PERCENT.

- [Overall Average Score](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=Overall%20Average%20Score>): Higher is better; unit: PERCENT.

- [Relative Average Performance](<https://www.sota2.com/research/sota/visual-grounding-on-refcoco-series.md?metric=Relative%20Average%20Performance>): Higher is better; unit: PERCENT.

## Recorded results

| Method | RefCOCO Test Score | Result date | Report paper |
| --- | --- | --- | --- |
| Vanilla | 77.73 | Not recorded | 6982d2dca8b01e76cc8e97fe |
| Nüwa | 68.17 | Not recorded | 6982d2dca8b01e76cc8e97fe |
| Nüwa | 38.67 | Not recorded | 6982d2dca8b01e76cc8e97fe |
| Nüwa | 18.67 | Not recorded | 6982d2dca8b01e76cc8e97fe |

### Vanilla

RefCOCO Test Score: 77.73. Result ID: 6997559c5698665ec4a7898b.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

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

Recorded evaluation configuration: {"Base Model":"LLaVA-Next 7B"}. This record does not assert independent reproduction.

### Nüwa

RefCOCO Test Score: 68.17. Result ID: 6997559c5698665ec4a7898e.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

Recorded evaluation configuration: {"Base Model":"LLaVA-Next 7B","Average Tokens":640}. This record does not assert independent reproduction.

### Nüwa

RefCOCO Test Score: 38.67. Result ID: 6997559c5698665ec4a7898d.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

Recorded evaluation configuration: {"Base Model":"LLaVA-Next 7B","Average Tokens":320}. This record does not assert independent reproduction.

### Nüwa

RefCOCO Test Score: 18.67. Result ID: 6997559c5698665ec4a7898c.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2dca8b01e76cc8e97fe.md>)

Recorded evaluation configuration: {"Base Model":"LLaVA-Next 7B","Average Tokens":160}. This record does not assert independent reproduction.

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