# Cycle Check on synthetic benchmark

Source: [Cycle Check on synthetic benchmark](<https://www.sota2.com/research/sota/cycle-check-on-synthetic-benchmark>)

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

Data updated: 2026-02-26T05:30:55.625Z

Evaluation of model performance on the Cycle Check task using the synthetic benchmark 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 | Cycle Check |
| Dataset (including any stated split) | synthetic benchmark |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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/cycle-check-on-synthetic-benchmark.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| NAG-LoRA | 99.9 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| Qwen3-LoRA | 99.8 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| NAG-Zero | 99.35 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| GraphTokenG | 97.95 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| GraphTokenE | 97.95 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| GraphTokenN | 96.92 | Not recorded | 69802f69a8b01e76cc8e14e4 |
| Qwen3-Direct | 76.2 | Not recorded | 69802f69a8b01e76cc8e14e4 |

### NAG-LoRA

Accuracy: 99.9. Result ID: 69965d0f5698665ec4a2c3f7.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### Qwen3-LoRA

Accuracy: 99.8. Result ID: 69965d0f5698665ec4a2c3f5.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### NAG-Zero

Accuracy: 99.35. Result ID: 69965d0f5698665ec4a2c3f6.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### GraphTokenG

Accuracy: 97.95. Result ID: 69965d0f5698665ec4a2c3f2.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72261e209131cea54caa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### GraphTokenE

Accuracy: 97.95. Result ID: 69965d0f5698665ec4a2c3f4.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72261e209131cea54caa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### GraphTokenN

Accuracy: 96.92. Result ID: 69965d0f5698665ec4a2c3f3.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72261e209131cea54caa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-600M"}. This record does not assert independent reproduction.

### Qwen3-Direct

Accuracy: 76.2. Result ID: 69965d0f5698665ec4a2c3f1.

[Reporting paper](<https://www.sota2.com/research/paper/69802f69a8b01e76cc8e14e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

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