# Fine-grained Image Classification on Life-Stages (LS)

Source: [Fine-grained Image Classification on Life-Stages (LS)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-life-stages-ls>)

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

Data updated: 2026-02-26T05:34:18.649Z

Evaluation of Fine-grained Image Classification performance on the Life-Stages (LS) dataset, measured using Accuracy.

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 | Fine-grained Image Classification |
| Dataset (including any stated split) | Life-Stages (LS) |
| Metric | Accuracy (LS) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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 (LS)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-life-stages-ls.md?metric=Accuracy%20(LS)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (LS) | Result date | Report paper |
| --- | --- | --- | --- |
| MetaCLIP-L | 64.3 | Not recorded | 68fa76b51e209131cea6e9a7 |
| OpenAI-B | 56.5 | Not recorded | 68fa76b51e209131cea6e9a7 |
| BT-Clip-O | 28 | Not recorded | 68fa76b51e209131cea6e9a7 |
| BT-Clip-M | 26.3 | Not recorded | 68fa76b51e209131cea6e9a7 |
| BioCLIP-B | 18 | Not recorded | 68fa76b51e209131cea6e9a7 |
| BT-Clip-B | 15 | Not recorded | 68fa76b51e209131cea6e9a7 |

### MetaCLIP-L

Accuracy (LS): 64.3. Result ID: 69944af83a6c3148491c50ca.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76721e209131cea6d0be.md>)

Recorded evaluation configuration: {"Backbone":"ViT-L-14","Pre-training":"MetaCLIP-cc","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### OpenAI-B

Accuracy (LS): 56.5. Result ID: 69944af83a6c3148491c50c8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

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

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

Recorded evaluation configuration: {"Backbone":"ViT-B-16","Pre-training":"OpenAI","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### BT-Clip-O

Accuracy (LS): 28. Result ID: 69944af83a6c3148491c50cb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-16","Pre-training":"Fine-tuned from OpenAI-B","Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction.

### BT-Clip-M

Accuracy (LS): 26.3. Result ID: 69944af83a6c3148491c50cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

Recorded evaluation configuration: {"Backbone":"ViT-L-14","Pre-training":"Fine-tuned from MetaCLIP-L","Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction.

### BioCLIP-B

Accuracy (LS): 18. Result ID: 69944af83a6c3148491c50c9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73361e209131cea5b2fc.md>)

[Implementation](<https://www.sota2.com/github.com/Imageomics/bioclip>)

Recorded evaluation configuration: {"Backbone":"ViT-B-16","Pre-training":"BioCLIP","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### BT-Clip-B

Accuracy (LS): 15. Result ID: 69944af83a6c3148491c50cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b51e209131cea6e9a7.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-16","Pre-training":"Fine-tuned from BioCLIP-B","Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction.

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