# Referring Expression Comprehension on RefCOCO 80 (val)

Source: [Referring Expression Comprehension on RefCOCO 80 (val)](<https://www.sota2.com/research/sota/referring-expression-comprehension-on-refcoco-80-val>)

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

Data updated: 2026-02-26T05:30:05.379Z

Evaluation of Referring Expression Comprehension performance on the validation split of the RefCOCO 80 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 | Referring Expression Comprehension |
| Dataset (including any stated split) | RefCOCO 80 (val) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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/referring-expression-comprehension-on-refcoco-80-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SimVG-TB | 92.99 | Not recorded | 68fa756f1e209131cea67e78 |
| SimVG-DB | 92.93 | Not recorded | 68fa756f1e209131cea67e78 |
| mPLUG-2 | 92.4 | Not recorded | 68fa756f1e209131cea67e78 |
| SimVG-DB | 91.47 | Not recorded | 68fa756f1e209131cea67e78 |
| SimVG-DB | 90.98 | Not recorded | 68fa756f1e209131cea67e78 |
| SimVG-TB | 90.59 | Not recorded | 68fa756f1e209131cea67e78 |
| PolyFormer | 90.38 | Not recorded | 68fa756f1e209131cea67e78 |
| OFA-L | 90.05 | Not recorded | 68fa756f1e209131cea67e78 |
| PolyFormer | 89.73 | Not recorded | 68fa756f1e209131cea67e78 |
| GroundingDINO | 89.19 | Not recorded | 68fa756f1e209131cea67e78 |
| DQ-DETR | 88.63 | Not recorded | 68fa756f1e209131cea67e78 |
| UniTAB | 88.59 | Not recorded | 68fa756f1e209131cea67e78 |
| SeqTR | 87 | Not recorded | 68fa756f1e209131cea67e78 |
| MDETR | 86.75 | Not recorded | 68fa756f1e209131cea67e78 |
| RefTR | 85.65 | Not recorded | 68fa756f1e209131cea67e78 |
| VILLA\_L | 82.39 | Not recorded | 68fa756f1e209131cea67e78 |
| UNITER\_L | 81.41 | Not recorded | 68fa756f1e209131cea67e78 |

### SimVG-TB

Accuracy: 92.99. Result ID: 6993be213a6c31484918376e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Implementation](<https://github.com/Dmmm1997/SimVG>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-L/32","Params (M)":"1.58","Pre-train images":"28K","Time (ms)":"101"}. This record does not assert independent reproduction.

### SimVG-DB

Accuracy: 92.93. Result ID: 6993be213a6c31484918376f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Implementation](<https://github.com/Dmmm1997/SimVG>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-L/32","Params (M)":"6.32","Pre-train images":"28K","Time (ms)":"116"}. This record does not assert independent reproduction.

### mPLUG-2

Accuracy: 92.4. Result ID: 6993be213a6c31484918376a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f461e209131cea44b05.md>)

[Implementation](<https://github.com/alibaba/AliceMind>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-L/14","Pre-train images":"14M"}. This record does not assert independent reproduction.

### SimVG-DB

Accuracy: 91.47. Result ID: 6993be213a6c31484918376d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Implementation](<https://github.com/Dmmm1997/SimVG>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-B/32","Params (M)":"6.32","Pre-train images":"174K","Time (ms)":"52"}. This record does not assert independent reproduction.

### SimVG-DB

Accuracy: 90.98. Result ID: 6993be213a6c31484918376b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Implementation](<https://github.com/Dmmm1997/SimVG>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-B/32","Params (M)":"6.32","Pre-train images":"28K","Time (ms)":"52"}. This record does not assert independent reproduction.

### SimVG-TB

Accuracy: 90.59. Result ID: 6993be213a6c31484918376c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Implementation](<https://github.com/Dmmm1997/SimVG>)

Recorded evaluation configuration: {"Visual Encoder":"ViT-B/32","Params (M)":"1.58","Pre-train images":"174K","Time (ms)":"44"}. This record does not assert independent reproduction.

### PolyFormer

Accuracy: 90.38. Result ID: 6993be213a6c314849183768.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428bb.md>)

[Implementation](<https://polyformer.github.io/>)

Recorded evaluation configuration: {"Visual Encoder":"Swin-L","Pre-train images":"174K"}. This record does not assert independent reproduction.

### OFA-L

Accuracy: 90.05. Result ID: 6993be213a6c314849183769.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35951.md>)

[Implementation](<https://github.com/OFA-Sys/OFA>)

Recorded evaluation configuration: {"Visual Encoder":"RN152","Pre-train images":"20M"}. This record does not assert independent reproduction.

### PolyFormer

Accuracy: 89.73. Result ID: 6993be213a6c314849183767.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428bb.md>)

[Implementation](<https://polyformer.github.io/>)

Recorded evaluation configuration: {"Visual Encoder":"Swin-B","Pre-train images":"174K"}. This record does not assert independent reproduction.

### GroundingDINO

Accuracy: 89.19. Result ID: 6993be213a6c314849183766.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

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

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

Recorded evaluation configuration: {"Visual Encoder":"Swin-T","Pre-train images":"200K","Time (ms)":"120"}. This record does not assert independent reproduction.

### DQ-DETR

Accuracy: 88.63. Result ID: 6993be213a6c314849183765.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1f1e209131cea3df1e.md>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Pre-train images":"200K"}. This record does not assert independent reproduction.

### UniTAB

Accuracy: 88.59. Result ID: 6993be213a6c314849183764.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d1f1e209131cea37ecb.md>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Pre-train images":"200K"}. This record does not assert independent reproduction.

### SeqTR

Accuracy: 87. Result ID: 6993be213a6c314849183763.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e0c1e209131cea3dad7.md>)

[Implementation](<https://github.com/sean-zhuh/SeqTR>)

Recorded evaluation configuration: {"Visual Encoder":"DN53","Params (M)":"7.90","Pre-train images":"174K","Time (ms)":"50"}. This record does not assert independent reproduction.

### MDETR

Accuracy: 86.75. Result ID: 6993be213a6c314849183761.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b5a9.md>)

[Implementation](<https://github.com/ashkamath/mdetr>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Params (M)":"17.36","Pre-train images":"200K","Time (ms)":"108"}. This record does not assert independent reproduction.

### RefTR

Accuracy: 85.65. Result ID: 6993be213a6c314849183762.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a611e209131cea27dec.md>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Params (M)":"17.86","Pre-train images":"100K","Time (ms)":"40"}. This record does not assert independent reproduction.

### VILLA\_L

Accuracy: 82.39. Result ID: 6993be213a6c314849183760.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d368.md>)

[Implementation](<https://github.com/zhegan27/VILLA>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Pre-train images":"4.6M"}. This record does not assert independent reproduction.

### UNITER\_L

Accuracy: 81.41. Result ID: 6993be213a6c31484918375f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67f91e209131cea1989f.md>)

[Implementation](<https://github.com/ChenRocks/UNITER>)

Recorded evaluation configuration: {"Visual Encoder":"RN101","Pre-train images":"4.6M"}. This record does not assert independent reproduction.

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