# Compositional Retrieval on ARO Retrieval

Source: [Compositional Retrieval on ARO Retrieval](<https://www.sota2.com/research/sota/compositional-retrieval-on-aro-retrieval>)

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

Data updated: 2026-02-26T05:29:39.999Z

Evaluation of model performance on the ARO Retrieval dataset for Compositional Retrieval tasks.

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 | Compositional Retrieval |
| Dataset (including any stated split) | ARO Retrieval |
| 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/compositional-retrieval-on-aro-retrieval.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CE-CLIP+ | 83.6 | Not recorded | 68fa730b1e209131cea5a166 |
| CE-CLIP | 83 | Not recorded | 68fa730b1e209131cea5a166 |
| CLIP-FT | 61.7 | Not recorded | 68fa730b1e209131cea5a166 |
| CLIP | 59.3 | Not recorded | 68fa730b1e209131cea5a166 |

### CE-CLIP+

Accuracy: 83.6. Result ID: 6992bcd23a6c3148491122d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Implementation](<https://github.com/lezhang7/Enhance-FineGrained>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### CE-CLIP

Accuracy: 83. Result ID: 6992bcd23a6c3148491122d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Implementation](<https://github.com/lezhang7/Enhance-FineGrained>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### CLIP-FT

Accuracy: 61.7. Result ID: 6992bcd23a6c3148491122d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned on COCO"}. This record does not assert independent reproduction.

### CLIP

Accuracy: 59.3. Result ID: 6992bcd23a6c3148491122cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

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