# Image Classification on CIFAR-100-LT (tau-based accuracies)

Source: [Image Classification on CIFAR-100-LT (tau-based accuracies)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-lt-tau-based-accuracies>)

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

Data updated: 2026-05-21T10:44:47.637Z

Evaluation of Image Classification performance on the CIFAR-100 Long Tail (LT) dataset, reporting accuracy metrics calculated under specific imbalance ratio thresholds (tau=0.1, 0.02, 0.01).

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 | Image Classification |
| Dataset (including any stated split) | CIFAR-100-LT |
| Metric | Accuracy (tau=0.1) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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 (tau=0.1)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-lt-tau-based-accuracies.md?metric=Accuracy%20(tau%3D0.1)>): Higher is better; unit: PERCENT.

- [Accuracy (tau=0.02)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-lt-tau-based-accuracies.md?metric=Accuracy%20(tau%3D0.02)>): Higher is better; unit: PERCENT.

- [Accuracy (tau=0.01)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-lt-tau-based-accuracies.md?metric=Accuracy%20(tau%3D0.01)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (tau=0.1) | Result date | Report paper |
| --- | --- | --- | --- |
| NONL | 57.4 | Not recorded | 6a0e9190f984739ed6738aa8 |
| NTCE | 56.8 | Not recorded | 6a0e9190f984739ed6738aa8 |
| CE | 55.5 | Not recorded | 6a0e9190f984739ed6738aa8 |
| NormFace | 54.6 | Not recorded | 6a0e9190f984739ed6738aa8 |
| ETF + DR | 54.4 | Not recorded | 6a0e9190f984739ed6738aa8 |

### NONL

Accuracy (tau=0.1): 57.4. Result ID: 6a0ee0a8fbc373592c06d4eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

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

### NTCE

Accuracy (tau=0.1): 56.8. Result ID: 6a0ee0a8fbc373592c06d4ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

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

### CE

Accuracy (tau=0.1): 55.5. Result ID: 6a0ee0a8fbc373592c06d4e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

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

### NormFace

Accuracy (tau=0.1): 54.6. Result ID: 6a0ee0a8fbc373592c06d4e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63811e209131ce9ff08e.md>)

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

### ETF + DR

Accuracy (tau=0.1): 54.4. Result ID: 6a0ee0a8fbc373592c06d4e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e9190f984739ed6738aa8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3ad3c.md>)

[Implementation](<https://www.sota2.com/link.>)

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

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