# CoT Soundness Evaluation on GSM8K

Source: [CoT Soundness Evaluation on GSM8K](<https://www.sota2.com/research/sota/cot-soundness-evaluation-on-gsm8k>)

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

Data updated: 2026-04-10T00:38:18.555Z

Evaluation of Chain-of-Thought (CoT) reasoning soundness on the GSM8K mathematical reasoning dataset, measured using the CSR metric.

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 | CoT Soundness Evaluation |
| Dataset (including any stated split) | GSM8K |
| Metric | CSR |
| 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

- [CSR](<https://www.sota2.com/research/sota/cot-soundness-evaluation-on-gsm8k.md?metric=CSR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CSR | Result date | Report paper |
| --- | --- | --- | --- |
| MirageBD | 95 | Not recorded | 69d732ba7d0172140ec76d19 |
| DecepChain | 43 | Not recorded | 69d732ba7d0172140ec76d19 |
| BadChain | 21 | Not recorded | 69d732ba7d0172140ec76d19 |

### MirageBD

CSR: 95. Result ID: 69d844eb47f34ce924c062ff.

[Reporting paper](<https://www.sota2.com/research/paper/69d732ba7d0172140ec76d19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732ba7d0172140ec76d19.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-1.5B","Trigger Activation":"true"}. This record does not assert independent reproduction.

### DecepChain

CSR: 43. Result ID: 69d844eb47f34ce924c062fe.

[Reporting paper](<https://www.sota2.com/research/paper/69d732ba7d0172140ec76d19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc31e209131cea8a40f.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-1.5B","Trigger Activation":"true"}. This record does not assert independent reproduction.

### BadChain

CSR: 21. Result ID: 69d844eb47f34ce924c062fd.

[Reporting paper](<https://www.sota2.com/research/paper/69d732ba7d0172140ec76d19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71fa1e209131cea53e33.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-1.5B","Trigger Activation":"true"}. This record does not assert independent reproduction.

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