# Risk-of-bias assessment in medical systematic reviews on COCHRANEFOREST (test)

Source: [Risk-of-bias assessment in medical systematic reviews on COCHRANEFOREST (test)](<https://www.sota2.com/research/sota/risk-of-bias-assessment-in-medical-systematic-reviews-on-cochraneforest-test>)

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

Data updated: 2026-02-26T05:39:29.919Z

Evaluation of model performance on the test split of the COCHRANEFOREST dataset for assessing risk-of-bias in medical systematic reviews, measured using Coherence and CA metrics.

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 | Risk-of-bias assessment in medical systematic reviews |
| Dataset (including any stated split) | COCHRANEFOREST (test) |
| Metric | Coherence |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Coherence](<https://www.sota2.com/research/sota/risk-of-bias-assessment-in-medical-systematic-reviews-on-cochraneforest-test.md?metric=Coherence>): Higher is better; unit: SCALAR.

- [CA](<https://www.sota2.com/research/sota/risk-of-bias-assessment-in-medical-systematic-reviews-on-cochraneforest-test.md?metric=CA>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Coherence | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2.5-7B-GRPO-VPRM | 0.895 | Not recorded | 697846b3a2b28e761d87c60e |
| Qwen2.5-7B-DAPO-VPRM | 0.801 | Not recorded | 697846b3a2b28e761d87c60e |
| Llama-3.1-405B | 0.507 | Not recorded | 697846b3a2b28e761d87c60e |
| Qwen2.5-72B | 0.443 | Not recorded | 697846b3a2b28e761d87c60e |
| GPT-OSS-120B | 0.362 | Not recorded | 697846b3a2b28e761d87c60e |

### Qwen2.5-7B-GRPO-VPRM

Coherence: 0.895. Result ID: 699670375698665ec4b7454e.

[Reporting paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

Recorded evaluation configuration: {"Training Algorithm":"GRPO","Reward Model":"VPRM"}. This record does not assert independent reproduction.

### Qwen2.5-7B-DAPO-VPRM

Coherence: 0.801. Result ID: 699670375698665ec4b7454f.

[Reporting paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

Recorded evaluation configuration: {"Training Algorithm":"DAPO","Reward Model":"VPRM"}. This record does not assert independent reproduction.

### Llama-3.1-405B

Coherence: 0.507. Result ID: 699670375698665ec4b7454d.

[Reporting paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Model Type":"Pretrained LLM"}. This record does not assert independent reproduction.

### Qwen2.5-72B

Coherence: 0.443. Result ID: 699670375698665ec4b7454c.

[Reporting paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Type":"Pretrained LLM"}. This record does not assert independent reproduction.

### GPT-OSS-120B

Coherence: 0.362. Result ID: 699670375698665ec4b7454b.

[Reporting paper](<https://www.sota2.com/research/paper/697846b3a2b28e761d87c60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ab01e209131cea84a0c.md>)

Recorded evaluation configuration: {"Model Type":"Pretrained LLM"}. This record does not assert independent reproduction.

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