# Class-Incremental Learning on ImageNet-A CIL T=10 (test)

Source: [Class-Incremental Learning on ImageNet-A CIL T=10 (test)](<https://www.sota2.com/research/sota/class-incremental-learning-on-imagenet-a-cil-t-10-test>)

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

Data updated: 2026-02-26T05:36:04.163Z

Evaluation of Class-Incremental Learning performance on the ImageNet-A CIL setup configured for 10 tasks, evaluated on the test split.

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 | Class-Incremental Learning |
| Dataset (including any stated split) | ImageNet-A CIL, T=10 (test) |
| Metric | Avg Accuracy |
| 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

- [Avg Accuracy](<https://www.sota2.com/research/sota/class-incremental-learning-on-imagenet-a-cil-t-10-test.md?metric=Avg%20Accuracy>): Higher is better; unit: PERCENT.

- [Rel. Error Rate](<https://www.sota2.com/research/sota/class-incremental-learning-on-imagenet-a-cil-t-10-test.md?metric=Rel.%20Error%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Avg Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| RanPAC | 62.4 | Not recorded | 68fa71cb1e209131cea53052 |
| RanPAC (No Phase 1) | 58.2 | Not recorded | 68fa71cb1e209131cea53052 |
| RanPAC (No RPs) | 57.7 | Not recorded | 68fa71cb1e209131cea53052 |
| Joint linear probe | 56.6 | Not recorded | 68fa71cb1e209131cea53052 |
| RanPAC (No RPs or Phase 1) | 54.4 | Not recorded | 68fa71cb1e209131cea53052 |
| ADaM | 52.6 | Not recorded | 68fa71cb1e209131cea53052 |
| RanPAC (NCM with Phase 1) | 49.7 | Not recorded | 68fa71cb1e209131cea53052 |
| RanPAC (NCM only) | 49.3 | Not recorded | 68fa71cb1e209131cea53052 |
| DualPrompt | 45.4 | Not recorded | 68fa71cb1e209131cea53052 |
| CODA-Prompt | 44.5 | Not recorded | 68fa71cb1e209131cea53052 |
| L2P | 42.5 | Not recorded | 68fa71cb1e209131cea53052 |

### RanPAC

Avg Accuracy: 62.4. Result ID: 6991fc683a6c3148490bdc0f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"Algorithm":1}. This record does not assert independent reproduction.

### RanPAC (No Phase 1)

Avg Accuracy: 58.2. Result ID: 6991fc683a6c3148490bdc11.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"Phase 1":"false"}. This record does not assert independent reproduction.

### RanPAC (No RPs)

Avg Accuracy: 57.7. Result ID: 6991fc683a6c3148490bdc10.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"RP layer":"false"}. This record does not assert independent reproduction.

### Joint linear probe

Avg Accuracy: 56.6. Result ID: 6991fc683a6c3148490bdc0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

Recorded evaluation configuration: {"Joint training":"true"}. This record does not assert independent reproduction.

### RanPAC (No RPs or Phase 1)

Avg Accuracy: 54.4. Result ID: 6991fc683a6c3148490bdc12.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"RP layer":"false","Phase 1":"false"}. This record does not assert independent reproduction.

### ADaM

Avg Accuracy: 52.6. Result ID: 6991fc683a6c3148490bdc0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74641e209131cea616e7.md>)

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

### RanPAC (NCM with Phase 1)

Avg Accuracy: 49.7. Result ID: 6991fc683a6c3148490bdc13.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"Evaluation":"NCM","Phase 1":"true"}. This record does not assert independent reproduction.

### RanPAC (NCM only)

Avg Accuracy: 49.3. Result ID: 6991fc683a6c3148490bdc14.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

Recorded evaluation configuration: {"Evaluation":"NCM","Phase 1":"false"}. This record does not assert independent reproduction.

### DualPrompt

Avg Accuracy: 45.4. Result ID: 6991fc683a6c3148490bdc0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

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

### CODA-Prompt

Avg Accuracy: 44.5. Result ID: 6991fc683a6c3148490bdc0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

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

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

### L2P

Avg Accuracy: 42.5. Result ID: 6991fc683a6c3148490bdc0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71cb1e209131cea53052.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71c71e209131cea52ec2.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.
