# Grounding and Counting on RefCOCOg (avg)

Source: [Grounding and Counting on RefCOCOg (avg)](<https://www.sota2.com/research/sota/grounding-and-counting-on-refcocog-avg>)

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

Data updated: 2026-03-19T09:05:31.405Z

Evaluation of combined Grounding and Counting performance on the average split of the RefCOCOg dataset.

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 | Grounding and Counting |
| Dataset (including any stated split) | RefCOCOg avg |
| Metric | Accuracy |
| 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](<https://www.sota2.com/research/sota/grounding-and-counting-on-refcocog-avg.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FineViT-VL | 88.81 | Not recorded | 69bb831c0636e5bf4afc1ac3 |
| Qwen3-VL | 85.43 | Not recorded | 69bb831c0636e5bf4afc1ac3 |
| Intern3.5-VL | 85.28 | Not recorded | 69bb831c0636e5bf4afc1ac3 |
| Aquila-VL | 58.71 | Not recorded | 69bb831c0636e5bf4afc1ac3 |

### FineViT-VL

Accuracy: 88.81. Result ID: 69bbbc5947f34ce924a5c969.

[Reporting paper](<https://www.sota2.com/research/paper/69bb831c0636e5bf4afc1ac3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bb831c0636e5bf4afc1ac3.md>)

Recorded evaluation configuration: {"Language Backbone":"Qwen3 1.7B LLM","Evaluation Framework":"VLMEvalKit \[69\]"}. This record does not assert independent reproduction.

### Qwen3-VL

Accuracy: 85.43. Result ID: 69bbbc5947f34ce924a5c96a.

[Reporting paper](<https://www.sota2.com/research/paper/69bb831c0636e5bf4afc1ac3.md>)

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

Recorded evaluation configuration: {"Language Backbone":"Qwen3 1.7B LLM","Evaluation Framework":"VLMEvalKit \[69\]"}. This record does not assert independent reproduction.

### Intern3.5-VL

Accuracy: 85.28. Result ID: 69bbbc5947f34ce924a5c96b.

[Reporting paper](<https://www.sota2.com/research/paper/69bb831c0636e5bf4afc1ac3.md>)

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

Recorded evaluation configuration: {"Language Backbone":"Qwen3 1.7B LLM","Evaluation Framework":"VLMEvalKit \[69\]"}. This record does not assert independent reproduction.

### Aquila-VL

Accuracy: 58.71. Result ID: 69bbbc5947f34ce924a5c96c.

[Reporting paper](<https://www.sota2.com/research/paper/69bb831c0636e5bf4afc1ac3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767c1e209131cea6d591.md>)

[Implementation](<https://github.com/LLaVA-VL/LLaVA-NeXT/tree/main/scripts/train>)

Recorded evaluation configuration: {"Language Backbone":"Qwen2.5 1.5B LLM","Evaluation Framework":"VLMEvalKit \[69\]"}. This record does not assert independent reproduction.

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