# Referring Expression Grounding on LVIS-Ref

Source: [Referring Expression Grounding on LVIS-Ref](<https://www.sota2.com/research/sota/referring-expression-grounding-on-lvis-ref>)

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

Data updated: 2026-02-26T05:34:11.263Z

Evaluation of Referring Expression Grounding performance on the LVIS-Ref dataset, typically measured using Accuracy for bounding boxes and points.

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) | LVIS-Ref |
| Metric | Accuracy (Box) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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 (Box)](<https://www.sota2.com/research/sota/referring-expression-grounding-on-lvis-ref.md?metric=Accuracy%20(Box)>): Higher is better; unit: PERCENT.

- [Accuracy (Point)](<https://www.sota2.com/research/sota/referring-expression-grounding-on-lvis-ref.md?metric=Accuracy%20(Point)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Box) | Result date | Report paper |
| --- | --- | --- | --- |
| Ferret-V2-13B | 87.7 | Not recorded | 68fa75051e209131cea65504 |
| Ferret-V2-7B | 86.6 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-30B | 84.9 | Not recorded | 68fa75051e209131cea65504 |
| Ferret-13B | 80.5 | Not recorded | 68fa75051e209131cea65504 |
| Ferret-7B | 79.4 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-7B | 79.4 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-3B-MoE | 79.3 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-3B | 76.3 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-1B-MoE | 71.4 | Not recorded | 68fa75051e209131cea65504 |
| MM1.5-1B | 69.7 | Not recorded | 68fa75051e209131cea65504 |
| Phi-3-Vision-4B | 53.8 | Not recorded | 68fa75051e209131cea65504 |
| MM1-30B | 53.4 | Not recorded | 68fa75051e209131cea65504 |
| MM1-7B | 53.1 | Not recorded | 68fa75051e209131cea65504 |
| MM1-3B | 52.9 | Not recorded | 68fa75051e209131cea65504 |
| MM1-1B | 51.4 | Not recorded | 68fa75051e209131cea65504 |
| LLaVA-OneVision-7B | 51.2 | Not recorded | 68fa75051e209131cea65504 |
| InternVL2 | 51 | Not recorded | 68fa75051e209131cea65504 |
| MiniCPM-v2-3B | 48.2 | Not recorded | 68fa75051e209131cea65504 |

### Ferret-V2-13B

Accuracy (Box): 87.7. Result ID: 699370673a6c31484915efef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72df1e209131cea5914a.md>)

Recorded evaluation configuration: {"Model Size Category":"Larger (\>13B)"}. This record does not assert independent reproduction.

### Ferret-V2-7B

Accuracy (Box): 86.6. Result ID: 699370673a6c31484915efeb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72df1e209131cea5914a.md>)

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

### MM1.5-30B

Accuracy (Box): 84.9. Result ID: 699370673a6c31484915eff1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"Larger (\>13B)"}. This record does not assert independent reproduction.

### Ferret-13B

Accuracy (Box): 80.5. Result ID: 699370673a6c31484915efee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d81e209131cea4da64.md>)

Recorded evaluation configuration: {"Model Size Category":"Larger (\>13B)"}. This record does not assert independent reproduction.

### Ferret-7B

Accuracy (Box): 79.4. Result ID: 699370673a6c31484915efea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d81e209131cea4da64.md>)

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

### MM1.5-7B

Accuracy (Box): 79.4. Result ID: 699370673a6c31484915efed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

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

### MM1.5-3B-MoE

Accuracy (Box): 79.3. Result ID: 699370673a6c31484915efe8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"3B","Architecture":"MoE"}. This record does not assert independent reproduction.

### MM1.5-3B

Accuracy (Box): 76.3. Result ID: 699370673a6c31484915efe7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"3B"}. This record does not assert independent reproduction.

### MM1.5-1B-MoE

Accuracy (Box): 71.4. Result ID: 699370673a6c31484915efe2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"1B","Architecture":"MoE"}. This record does not assert independent reproduction.

### MM1.5-1B

Accuracy (Box): 69.7. Result ID: 699370673a6c31484915efe1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"1B"}. This record does not assert independent reproduction.

### Phi-3-Vision-4B

Accuracy (Box): 53.8. Result ID: 699370673a6c31484915efe4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e1.md>)

Recorded evaluation configuration: {"Model Size Category":"3B"}. This record does not assert independent reproduction.

### MM1-30B

Accuracy (Box): 53.4. Result ID: 699370673a6c31484915eff0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea594e3.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"Larger (\>13B)"}. This record does not assert independent reproduction.

### MM1-7B

Accuracy (Box): 53.1. Result ID: 699370673a6c31484915efec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea594e3.md>)

[Implementation](<https://github.com/apple/axlearn>)

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

### MM1-3B

Accuracy (Box): 52.9. Result ID: 699370673a6c31484915efe6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea594e3.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"3B"}. This record does not assert independent reproduction.

### MM1-1B

Accuracy (Box): 51.4. Result ID: 699370673a6c31484915efe0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea594e3.md>)

[Implementation](<https://github.com/apple/axlearn>)

Recorded evaluation configuration: {"Model Size Category":"1B"}. This record does not assert independent reproduction.

### LLaVA-OneVision-7B

Accuracy (Box): 51.2. Result ID: 699370673a6c31484915efe9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e1.md>)

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

### InternVL2

Accuracy (Box): 51. Result ID: 699370673a6c31484915efe5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73081e209131cea59f8b.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Model Size Category":"3B"}. This record does not assert independent reproduction.

### MiniCPM-v2-3B

Accuracy (Box): 48.2. Result ID: 699370673a6c31484915efe3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65504.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74681e209131cea6183d.md>)

[Implementation](<https://github.com/OpenBMB/MiniCPM-V>)

Recorded evaluation configuration: {"Model Size Category":"3B"}. This record does not assert independent reproduction.

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