# Domain Generalization on Waterbirds

Source: [Domain Generalization on Waterbirds](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds>)

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

Data updated: 2026-02-26T05:54:05.248Z

Evaluation of model generalization performance on the Waterbirds dataset, typically measured using Out-of-Distribution (OOD) accuracy.

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 | Domain Generalization |
| Dataset (including any stated split) | Waterbirds |
| Metric | OOD Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [OOD Accuracy](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds.md?metric=OOD%20Accuracy>): Higher is better; unit: PERCENT.

- [OOD Accuracy (LL)](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds.md?metric=OOD%20Accuracy%20(LL)>): Higher is better; unit: PERCENT.

- [OOD Accuracy (LW)](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds.md?metric=OOD%20Accuracy%20(LW)>): Higher is better; unit: PERCENT.

- [OOD Accuracy (WL)](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds.md?metric=OOD%20Accuracy%20(WL)>): Higher is better; unit: PERCENT.

- [OOD Accuracy (WW)](<https://www.sota2.com/research/sota/domain-generalization-on-waterbirds.md?metric=OOD%20Accuracy%20(WW)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | OOD Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Sparse and Shared Feature Activations | 90.5 | Not recorded | 68fa717d1e209131cea5196d |
| Sparse and Shared Feature Activations | 89.2 | Not recorded | 68fa717d1e209131cea5196d |
| Sparse and Shared Feature Activations | 88.4 | Not recorded | 68fa717d1e209131cea5196d |
| Sparse and Shared Feature Activations | 87.6 | Not recorded | 68fa717d1e209131cea5196d |
| ERM | 84.2 | Not recorded | 68fa717d1e209131cea5196d |
| ERM | 81.3 | Not recorded | 68fa717d1e209131cea5196d |
| ERM | 79.8 | Not recorded | 68fa717d1e209131cea5196d |
| ERM | 79.8 | Not recorded | 68fa717d1e209131cea5196d |

### Sparse and Shared Feature Activations

OOD Accuracy: 90.5. Result ID: 6991e4b13a6c3148490b46b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

Recorded evaluation configuration: {"Framework":"DomainBed"}. This record does not assert independent reproduction.

### Sparse and Shared Feature Activations

OOD Accuracy: 89.2. Result ID: 6991e4b13a6c3148490b46ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

Recorded evaluation configuration: {"N-shot":"10"}. This record does not assert independent reproduction.

### Sparse and Shared Feature Activations

OOD Accuracy: 88.4. Result ID: 6991e4b13a6c3148490b46a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

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

### Sparse and Shared Feature Activations

OOD Accuracy: 87.6. Result ID: 6991e4b13a6c3148490b46aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

Recorded evaluation configuration: {"N-shot":"5"}. This record does not assert independent reproduction.

### ERM

OOD Accuracy: 84.2. Result ID: 6991e4b13a6c3148490b46ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68bd1e209131cea1dec5.md>)

Recorded evaluation configuration: {"N-shot":"10"}. This record does not assert independent reproduction.

### ERM

OOD Accuracy: 81.3. Result ID: 6991e4b13a6c3148490b46b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68bd1e209131cea1dec5.md>)

Recorded evaluation configuration: {"Framework":"DomainBed"}. This record does not assert independent reproduction.

### ERM

OOD Accuracy: 79.8. Result ID: 6991e4b13a6c3148490b46a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68bd1e209131cea1dec5.md>)

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

### ERM

OOD Accuracy: 79.8. Result ID: 6991e4b13a6c3148490b46a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea5196d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68bd1e209131cea1dec5.md>)

Recorded evaluation configuration: {"N-shot":"5"}. This record does not assert independent reproduction.

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