# Systematic Generalization on SCAN Around Right (test)

Source: [Systematic Generalization on SCAN Around Right (test)](<https://www.sota2.com/research/sota/systematic-generalization-on-scan-around-right-test>)

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

Data updated: 2026-02-27T08:32:31.939Z

Evaluation of Systematic Generalization performance on the test split of the SCAN Around Right dataset, measured by 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 | Systematic Generalization |
| Dataset (including any stated split) | SCAN Around Right (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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](<https://www.sota2.com/research/sota/systematic-generalization-on-scan-around-right-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| EquiLSTM | 95.7 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiZeroLSTM | 92.5 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiGRU | 91.6 | Not recorded | 68fa70d71e209131cea4d957 |
| G-LSTM | 89.6 | Not recorded | 68fa70d71e209131cea4d957 |
| G-RNN | 84.5 | Not recorded | 68fa70d71e209131cea4d957 |
| G-GRU | 82.7 | Not recorded | 68fa70d71e209131cea4d957 |
| EquizeroLSTM | 81.7 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiRNN | 78.4 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiZeroGRU | 74.6 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiZeroRNN | 64.3 | Not recorded | 68fa70d71e209131cea4d957 |
| EquiLSTM | 28.2 | Not recorded | 68fa70d71e209131cea4d957 |
| RNN | 5.9 | Not recorded | 68fa70d71e209131cea4d957 |
| LSTM | 0.4 | Not recorded | 68fa70d71e209131cea4d957 |
| LSTM | 0.4 | Not recorded | 68fa70d71e209131cea4d957 |
| GRU | 0.1 | Not recorded | 68fa70d71e209131cea4d957 |

### EquiLSTM

Accuracy: 95.7. Result ID: 69a1568935dfcabb00e65cff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Training Iterations":"10K","Evaluation Protocol":"fine-tuned"}. This record does not assert independent reproduction.

### EquiZeroLSTM

Accuracy: 92.5. Result ID: 69a1568935dfcabb00e65d00.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Training Iterations":"10K","Evaluation Protocol":"fine-tuned"}. This record does not assert independent reproduction.

### EquiGRU

Accuracy: 91.6. Result ID: 69a1568935dfcabb00e65d0f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Fine-tuning iterations":"10K"}. This record does not assert independent reproduction.

### G-LSTM

Accuracy: 89.6. Result ID: 69a1568935dfcabb00e65cfe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"Dir.","Training Iterations":"200K","Evaluation Protocol":"trained"}. This record does not assert independent reproduction.

### G-RNN

Accuracy: 84.5. Result ID: 69a1568935dfcabb00e65ce2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"Dir.","Training iterations":"200K iterations"}. This record does not assert independent reproduction.

### G-GRU

Accuracy: 82.7. Result ID: 69a1568935dfcabb00e65d0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"Dir.","Training iterations":"200K"}. This record does not assert independent reproduction.

### EquizeroLSTM

Accuracy: 81.7. Result ID: 69a1568935dfcabb00e65d28.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Direction","Zero-shot":"true","Iterations":"200K"}. This record does not assert independent reproduction.

### EquiRNN

Accuracy: 78.4. Result ID: 69a1568935dfcabb00e65ce3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Training iterations":"10K iterations fine-tuning"}. This record does not assert independent reproduction.

### EquiZeroGRU

Accuracy: 74.6. Result ID: 69a1568935dfcabb00e65d10.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Fine-tuning iterations":"10K"}. This record does not assert independent reproduction.

### EquiZeroRNN

Accuracy: 64.3. Result ID: 69a1568935dfcabb00e65ce4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Dir.","Training iterations":"10K iterations fine-tuning"}. This record does not assert independent reproduction.

### EquiLSTM

Accuracy: 28.2. Result ID: 69a1568935dfcabb00e65d27.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

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

Recorded evaluation configuration: {"Group":"Direction","Zero-shot":"true","Iterations":"200K"}. This record does not assert independent reproduction.

### RNN

Accuracy: 5.9. Result ID: 69a1568935dfcabb00e65ce1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"-","Training iterations":"200K iterations"}. This record does not assert independent reproduction.

### LSTM

Accuracy: 0.4. Result ID: 69a1568935dfcabb00e65cfd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"–","Training Iterations":"200K","Evaluation Protocol":"trained"}. This record does not assert independent reproduction.

### LSTM

Accuracy: 0.4. Result ID: 69a1568935dfcabb00e65d26.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"–","Zero-shot":"true","Iterations":"200K"}. This record does not assert independent reproduction.

### GRU

Accuracy: 0.1. Result ID: 69a1568935dfcabb00e65d0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d957.md>)

Recorded evaluation configuration: {"Group":"–","Training iterations":"200K"}. This record does not assert independent reproduction.

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