# Clinical Diagnostic Reasoning on MedCaseReasoning 50 randomly sampled (test)

Source: [Clinical Diagnostic Reasoning on MedCaseReasoning 50 randomly sampled (test)](<https://www.sota2.com/research/sota/clinical-diagnostic-reasoning-on-medcasereasoning-50-randomly-sampled-test>)

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

Data updated: 2026-04-14T11:22:35.711Z

Evaluation of model performance on the 50 randomly sampled subset of the MedCaseReasoning test set, focusing on Clinical Diagnostic Reasoning capabilities.

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 | Clinical Diagnostic Reasoning |
| Dataset (including any stated split) | MedCaseReasoning 50 randomly sampled (test) |
| Metric | Clinician Trust Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Clinician Trust Score](<https://www.sota2.com/research/sota/clinical-diagnostic-reasoning-on-medcasereasoning-50-randomly-sampled-test.md?metric=Clinician%20Trust%20Score>): Higher is better; unit: SCALAR.

- [Dx Accuracy](<https://www.sota2.com/research/sota/clinical-diagnostic-reasoning-on-medcasereasoning-50-randomly-sampled-test.md?metric=Dx%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Clinician Trust Score | Result date | Report paper |
| --- | --- | --- | --- |
| CGCL | 71 | Not recorded | 69ddca6c7d0172140ec79edb |
| GRPO | 61 | Not recorded | 69ddca6c7d0172140ec79edb |
| SFT-CoT | 54.5 | Not recorded | 69ddca6c7d0172140ec79edb |
| Trust-Think & Answer | 51.5 | Not recorded | 69ddca6c7d0172140ec79edb |

### CGCL

Clinician Trust Score: 71. Result ID: 69de237047f34ce924c69559.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca6c7d0172140ec79edb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca6c7d0172140ec79edb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GRPO

Clinician Trust Score: 61. Result ID: 69de237047f34ce924c69558.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca6c7d0172140ec79edb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SFT-CoT

Clinician Trust Score: 54.5. Result ID: 69de237047f34ce924c69557.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca6c7d0172140ec79edb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Trust-Think & Answer

Clinician Trust Score: 51.5. Result ID: 69de237047f34ce924c69556.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca6c7d0172140ec79edb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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