# Long-tail Class Incremental Learning on ImageNet-LT 20 tasks 1.0

Source: [Long-tail Class Incremental Learning on ImageNet-LT 20 tasks 1.0](<https://www.sota2.com/research/sota/long-tail-class-incremental-learning-on-imagenet-lt-20-tasks-1-0>)

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

Data updated: 2026-03-24T16:04:40.607Z

Evaluation of Long-tail Class Incremental Learning performance on the ImageNet-LT dataset configured for 20 tasks (version 1.0).

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 | Long-tail Class Incremental Learning |
| Dataset (including any stated split) | ImageNet-LT 20 tasks 1.0 |
| 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/long-tail-class-incremental-learning-on-imagenet-lt-20-tasks-1-0.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Upper Bound | 75.6 | Not recorded | 69c21aa60636e5bf4afc4a79 |
| Ours | 68.2 | Not recorded | 69c21aa60636e5bf4afc4a79 |
| CLIP Zero-Shot | 63.1 | Not recorded | 69c21aa60636e5bf4afc4a79 |
| Adapter Fine-tuning | 19.6 | Not recorded | 69c21aa60636e5bf4afc4a79 |

### Upper Bound

Accuracy: 75.6. Result ID: 69c2b60547f34ce924ac033e.

[Reporting paper](<https://www.sota2.com/research/paper/69c21aa60636e5bf4afc4a79.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Evaluation Protocol":"Joint Learning"}. This record does not assert independent reproduction.

### Ours

Accuracy: 68.2. Result ID: 69c2b60547f34ce924ac0341.

[Reporting paper](<https://www.sota2.com/research/paper/69c21aa60636e5bf4afc4a79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21aa60636e5bf4afc4a79.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16"}. This record does not assert independent reproduction.

### CLIP Zero-Shot

Accuracy: 63.1. Result ID: 69c2b60547f34ce924ac033f.

[Reporting paper](<https://www.sota2.com/research/paper/69c21aa60636e5bf4afc4a79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8e1e209131cea3a917.md>)

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

### Adapter Fine-tuning

Accuracy: 19.6. Result ID: 69c2b60547f34ce924ac0340.

[Reporting paper](<https://www.sota2.com/research/paper/69c21aa60636e5bf4afc4a79.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16"}. This record does not assert independent reproduction.

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