# Few-Shot Class-Incremental Learning on ImageNet-1K 500 base classes 2009

Source: [Few-Shot Class-Incremental Learning on ImageNet-1K 500 base classes 2009](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009>)

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

Data updated: 2026-02-26T05:34:26.228Z

Evaluation of Few-Shot Class-Incremental Learning performance on the ImageNet-1K subset defined by 500 base classes from the 2009 configuration, tracking accuracy across sequential learning sessions.

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 | Few-Shot Class-Incremental Learning |
| Dataset (including any stated split) | ImageNet-1K 500 base classes 2009 |
| Metric | Accuracy (Session 0) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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 (Session 0)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%200)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 1)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%201)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 2)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%202)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 3)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%203)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 4)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%204)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 5)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%205)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 6)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%206)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 7)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%207)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 8)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%208)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 9)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%209)>): Higher is better; unit: PERCENT.

- [Accuracy (Session 10)](<https://www.sota2.com/research/sota/few-shot-class-incremental-learning-on-imagenet-1k-500-base-classes-2009.md?metric=Accuracy%20(Session%2010)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Session 0) | Result date | Report paper |
| --- | --- | --- | --- |
| Ours | 39.34 | Not recorded | 68fa74c61e209131cea63b8c |
| SAVC | 37.74 | Not recorded | 68fa74c61e209131cea63b8c |
| SoftNet | 36.21 | Not recorded | 68fa74c61e209131cea63b8c |

### Ours

Accuracy (Session 0): 39.34. Result ID: 699367063a6c31484915ae0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74c61e209131cea63b8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74c61e209131cea63b8c.md>)

Recorded evaluation configuration: {"Way":"50-way","Shot":"10-shot"}. This record does not assert independent reproduction.

### SAVC

Accuracy (Session 0): 37.74. Result ID: 699367063a6c31484915ae0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74c61e209131cea63b8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ef31e209131cea431f0.md>)

[Implementation](<https://github.com/zysong0113/SAVC>)

Recorded evaluation configuration: {"Way":"50-way","Shot":"10-shot"}. This record does not assert independent reproduction.

### SoftNet

Accuracy (Session 0): 36.21. Result ID: 699367063a6c31484915ae0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74c61e209131cea63b8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea71e209131cea411c6.md>)

Recorded evaluation configuration: {"Way":"50-way","Shot":"10-shot"}. This record does not assert independent reproduction.

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