# Novel Class Discovery on ImageNet-100 (test)

Source: [Novel Class Discovery on ImageNet-100 (test)](<https://www.sota2.com/research/sota/novel-class-discovery-on-imagenet-100-test>)

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

Data updated: 2026-02-26T06:04:10.249Z

Evaluation of Novel Class Discovery performance on the test split of the ImageNet-100 dataset, measured using Seen, Novel, and All Accuracies.

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 | Novel Class Discovery |
| Dataset (including any stated split) | ImageNet-100 (test) |
| Metric | Seen Accuracy |
| 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

- [Seen Accuracy](<https://www.sota2.com/research/sota/novel-class-discovery-on-imagenet-100-test.md?metric=Seen%20Accuracy>): Higher is better; unit: PERCENT.

- [Novel Accuracy](<https://www.sota2.com/research/sota/novel-class-discovery-on-imagenet-100-test.md?metric=Novel%20Accuracy>): Higher is better; unit: PERCENT.

- [All Accuracy](<https://www.sota2.com/research/sota/novel-class-discovery-on-imagenet-100-test.md?metric=All%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Seen Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| OwMatch+ | 87.8 | Not recorded | 68fa75851e209131cea68592 |
| OwMatch | 86.4 | Not recorded | 68fa75851e209131cea68592 |
| ORCA | 83.9 | Not recorded | 68fa75851e209131cea68592 |
| TRSSL | 82.6 | Not recorded | 68fa75851e209131cea68592 |
| UNO | 66 | Not recorded | 68fa75851e209131cea68592 |

### OwMatch+

Seen Accuracy: 87.8. Result ID: 6993b9f93a6c3148491817fc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Implementation](<https://github.com/niusj03/OwMatch>)

Recorded evaluation configuration: {"Novel class ratio":"50%","Labeled ratio":"10%"}. This record does not assert independent reproduction.

### OwMatch

Seen Accuracy: 86.4. Result ID: 6993b9f93a6c3148491817fb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Implementation](<https://github.com/niusj03/OwMatch>)

Recorded evaluation configuration: {"Novel class ratio":"50%","Labeled ratio":"10%"}. This record does not assert independent reproduction.

### ORCA

Seen Accuracy: 83.9. Result ID: 6993b9f93a6c3148491817f9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ba21e209131cea2f704.md>)

[Implementation](<https://github.com/snap-stanford/orca>)

Recorded evaluation configuration: {"Novel class ratio":"50%","Labeled ratio":"10%"}. This record does not assert independent reproduction.

### TRSSL

Seen Accuracy: 82.6. Result ID: 6993b9f93a6c3148491817fa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d201e209131cea37fac.md>)

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

Recorded evaluation configuration: {"Novel class ratio":"50%","Labeled ratio":"10%"}. This record does not assert independent reproduction.

### UNO

Seen Accuracy: 66. Result ID: 6993b9f93a6c3148491817f8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75851e209131cea68592.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ae91e209131cea2afa3.md>)

[Implementation](<https://ncd-uno.github.io/>)

Recorded evaluation configuration: {"Novel class ratio":"50%","Labeled ratio":"10%"}. This record does not assert independent reproduction.

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