# Machine Unlearning on Tiny-ImageNet Forget 10 Class

Source: [Machine Unlearning on Tiny-ImageNet Forget 10 Class](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class>)

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

Data updated: 2026-06-01T07:46:56.079Z

Evaluation of Machine Unlearning performance on the Tiny-ImageNet dataset, specifically targeting the forgetting of 10 classes. Metrics focus on retention, forgetting accuracy, and unlearning resistance.

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 | Machine Unlearning |
| Dataset (including any stated split) | Tiny-ImageNet Forget 10 Class |
| Metric | Accuracy (Forgetting) |
| 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 (Forgetting)](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class.md?metric=Accuracy%20(Forgetting)>): Higher is better; unit: PERCENT.

- [Accuracy (Retention)](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class.md?metric=Accuracy%20(Retention)>): Higher is better; unit: PERCENT.

- [Overlap Uncertainty (epsilon)](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class.md?metric=Overlap%20Uncertainty%20(epsilon)>): Lower is better; unit: PERCENT.

- [Prototype Accuracy (Forgetting)](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class.md?metric=Prototype%20Accuracy%20(Forgetting)>): Higher is better; unit: PERCENT.

- [Robust Accuracy (Retention)](<https://www.sota2.com/research/sota/machine-unlearning-on-tiny-imagenet-forget-10-class.md?metric=Robust%20Accuracy%20(Retention)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Forgetting) | Result date | Report paper |
| --- | --- | --- | --- |
| Original Model | 99.68 | Not recorded | 68fa79171e209131cea7c144 |
| Spotter | 0.56 | Not recorded | 68fa79171e209131cea7c144 |
| DELETE | 0.44 | Not recorded | 68fa79171e209131cea7c144 |
| Retrain Model | 0 | Not recorded | 68fa79171e209131cea7c144 |

### Original Model

Accuracy (Forgetting): 99.68. Result ID: 6a1d38defbc373592c161e8e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79171e209131cea7c144.md>)

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

### Spotter

Accuracy (Forgetting): 0.56. Result ID: 6a1d38defbc373592c161e91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79171e209131cea7c144.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79171e209131cea7c144.md>)

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

### DELETE

Accuracy (Forgetting): 0.44. Result ID: 6a1d38defbc373592c161e90.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79171e209131cea7c144.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ae1e209131cea74ad5.md>)

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

### Retrain Model

Accuracy (Forgetting): 0. Result ID: 6a1d38defbc373592c161e8f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79171e209131cea7c144.md>)

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.