# Continual Learning Performance Metrics (A10, Avg.) on ImageNet-A (test)

Source: [Continual Learning Performance Metrics (A10, Avg.) on ImageNet-A (test)](<https://www.sota2.com/research/sota/continual-learning-performance-metrics-a10-avg-on-imagenet-a-test>)

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

Data updated: 2026-02-26T05:22:13.768Z

Evaluation of Continual Learning performance on the test split of the ImageNet-A dataset, specifically reporting metrics A10 and Average performance.

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 | Continual Learning |
| Dataset (including any stated split) | ImageNet-A (test) |
| Metric | A10 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [A10](<https://www.sota2.com/research/sota/continual-learning-performance-metrics-a10-avg-on-imagenet-a-test.md?metric=A10>): Higher is better; unit: PERCENT.

- [Average Performance](<https://www.sota2.com/research/sota/continual-learning-performance-metrics-a10-avg-on-imagenet-a-test.md?metric=Average%20Performance>): Higher is better; unit: PERCENT.

## Recorded results

| Method | A10 | Result date | Report paper |
| --- | --- | --- | --- |
| Joint Train | 65.01 | Not recorded | 6997eabd09c33900fcd7679a |
| EWC-LoRA | 59.89 | Not recorded | 6997eabd09c33900fcd7679a |
| CL-LoRA | 57.62 | Not recorded | 6997eabd09c33900fcd7679a |
| SD-LoRA | 55.23 | Not recorded | 6997eabd09c33900fcd7679a |
| BiLoRA | 51.05 | Not recorded | 6997eabd09c33900fcd7679a |
| InfLoRA | 50.75 | Not recorded | 6997eabd09c33900fcd7679a |
| DualPrompt | 45.49 | Not recorded | 6997eabd09c33900fcd7679a |
| CODA-Prompt | 45.36 | Not recorded | 6997eabd09c33900fcd7679a |
| L2P | 42.94 | Not recorded | 6997eabd09c33900fcd7679a |
| Vanilla LoRA | 40.01 | Not recorded | 6997eabd09c33900fcd7679a |
| Finetune | 32.85 | Not recorded | 6997eabd09c33900fcd7679a |

### Joint Train

A10: 65.01. Result ID: 699b0f245698665ec4a259ff.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Joint"}. This record does not assert independent reproduction.

### EWC-LoRA

A10: 59.89. Result ID: 699b0f245698665ec4a25a09.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### CL-LoRA

A10: 57.62. Result ID: 699b0f245698665ec4a25a06.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79041e209131cea7bb1b.md>)

[Implementation](<https://github.com/JiangpengHe/CL-LoRA>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### SD-LoRA

A10: 55.23. Result ID: 699b0f245698665ec4a25a05.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77461e209131cea7225e.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### BiLoRA

A10: 51.05. Result ID: 699b0f245698665ec4a25a07.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### InfLoRA

A10: 50.75. Result ID: 699b0f245698665ec4a25a04.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72bb1e209131cea58354.md>)

[Implementation](<https://github.com/liangyanshuo/InfLoRA>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### DualPrompt

A10: 45.49. Result ID: 699b0f245698665ec4a25a02.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d341e209131cea38614.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### CODA-Prompt

A10: 45.36. Result ID: 699b0f245698665ec4a25a03.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee41e209131cea42add.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### L2P

A10: 42.94. Result ID: 699b0f245698665ec4a25a01.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c0c1e209131cea31ad7.md>)

[Implementation](<https://github.com/google-research/l2p>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### Vanilla LoRA

A10: 40.01. Result ID: 699b0f245698665ec4a25a08.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

### Finetune

A10: 32.85. Result ID: 699b0f245698665ec4a25a00.

[Reporting paper](<https://www.sota2.com/research/paper/6997eabd09c33900fcd7679a.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Pre-trained":"ImageNet-21K","Training Strategy":"Sequential"}. This record does not assert independent reproduction.

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