# Joint image-text clustering on ImageNet 200 classes (val)

Source: [Joint image-text clustering on ImageNet 200 classes (val)](<https://www.sota2.com/research/sota/joint-image-text-clustering-on-imagenet-200-classes-val>)

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

Data updated: 2026-04-02T08:12:57.203Z

Evaluation of joint image-text clustering performance on the validation split of the ImageNet 200 classes dataset, measured using GD, V-Measure, and ARI.

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 | Joint image-text clustering |
| Dataset (including any stated split) | ImageNet 200 classes (val) |
| Metric | GD |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [GD](<https://www.sota2.com/research/sota/joint-image-text-clustering-on-imagenet-200-classes-val.md?metric=GD>): Higher is better; unit: SCALAR.

- [V-Measure](<https://www.sota2.com/research/sota/joint-image-text-clustering-on-imagenet-200-classes-val.md?metric=V-Measure>): Higher is better; unit: SCALAR.

- [ARI](<https://www.sota2.com/research/sota/joint-image-text-clustering-on-imagenet-200-classes-val.md?metric=ARI>): Higher is better; unit: SCALAR.

## Recorded results

| Method | GD | Result date | Report paper |
| --- | --- | --- | --- |
| Original CLIP | 0.685 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| Mean-Centering | 0.685 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| CLIP-Refine | 0.647 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| M2-Mix | 0.637 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| CS-Aligner | 0.617 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.61 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.554 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| AlignCLIP | 0.55 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.474 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.441 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.417 | Not recorded | 69cdf7f70636e5bf4afcc9ea |
| TPC-CMA | 0.392 | Not recorded | 69cdf7f70636e5bf4afcc9ea |

### Original CLIP

GD: 0.685. Result ID: 69ce24bb47f34ce924b73678.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

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

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Mean-Centering

GD: 0.685. Result ID: 69ce24bb47f34ce924b73679.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dbc1e209131cea3bb74.md>)

[Implementation](<https://github.com/Weixin-Liang/Modality-Gap>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CLIP-Refine

GD: 0.647. Result ID: 69ce24bb47f34ce924b7367c.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76f8d.md>)

[Implementation](<https://github.com/yshinya6/clip-refine>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### M2-Mix

GD: 0.637. Result ID: 69ce24bb47f34ce924b7367b.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71241e209131cea4f76e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CS-Aligner

GD: 0.617. Result ID: 69ce24bb47f34ce924b7367d.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a41e209131cea7988d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.61. Result ID: 69ce24bb47f34ce924b7367e.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.01"}. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.554. Result ID: 69ce24bb47f34ce924b7367f.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.05"}. This record does not assert independent reproduction.

### AlignCLIP

GD: 0.55. Result ID: 69ce24bb47f34ce924b7367a.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.474. Result ID: 69ce24bb47f34ce924b73680.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.3"}. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.441. Result ID: 69ce24bb47f34ce924b73681.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.5"}. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.417. Result ID: 69ce24bb47f34ce924b73682.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.7"}. This record does not assert independent reproduction.

### TPC-CMA

GD: 0.392. Result ID: 69ce24bb47f34ce924b73683.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f70636e5bf4afcc9ea.md>)

Recorded evaluation configuration: {"alpha\_target":"0.9"}. This record does not assert independent reproduction.

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