# Medical Report Generation on MIMIC-CXR (Metrics: ROUGE-L, CIDEr, SemScore)

Source: [Medical Report Generation on MIMIC-CXR (Metrics: ROUGE-L, CIDEr, SemScore)](<https://www.sota2.com/research/sota/medical-report-generation-on-mimic-cxr-metrics-rouge-l-cider-semscore>)

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

Data updated: 2026-02-26T06:00:44.070Z

Evaluation of models generating medical reports from the MIMIC-CXR dataset, specifically measured using ROUGE-L, CIDEr, and SemScore.

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 | Medical Report Generation |
| Dataset (including any stated split) | MIMIC-CXR |
| Metric | ROUGE-L |
| 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

- [ROUGE-L](<https://www.sota2.com/research/sota/medical-report-generation-on-mimic-cxr-metrics-rouge-l-cider-semscore.md?metric=ROUGE-L>): Higher is better; unit: PERCENT.

- [CIDEr](<https://www.sota2.com/research/sota/medical-report-generation-on-mimic-cxr-metrics-rouge-l-cider-semscore.md?metric=CIDEr>): Higher is better; unit: PERCENT.

- [SemScore](<https://www.sota2.com/research/sota/medical-report-generation-on-mimic-cxr-metrics-rouge-l-cider-semscore.md?metric=SemScore>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ROUGE-L | Result date | Report paper |
| --- | --- | --- | --- |
| Lingshu-7B + Med-Scout | 31.4 | Not recorded | 69802fdda8b01e76cc8e185a |
| Lingshu-7B | 30.9 | Not recorded | 69802fdda8b01e76cc8e185a |
| MedGemma-4B-IT | 27.4 | Not recorded | 69802fdda8b01e76cc8e185a |
| HuatuoGPT-Vision-7B + Med-Scout | 25.7 | Not recorded | 69802fdda8b01e76cc8e185a |
| GEMINI-3-FLASH | 24.9 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen3-VL-8B-Instruct + Med-Scout | 23.8 | Not recorded | 69802fdda8b01e76cc8e185a |
| HuatuoGPT-Vision-7B | 23.6 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen3-VL-4B-Instruct + Med-Scout | 23.4 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen2.5-VL-7B-Instruct | 22.9 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen3-VL-4B-Instruct | 21.8 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen2.5-VL-3B-Instruct | 21.4 | Not recorded | 69802fdda8b01e76cc8e185a |
| Qwen3-VL-8B-Instruct | 21.3 | Not recorded | 69802fdda8b01e76cc8e185a |
| InternVL3-8B | 20.7 | Not recorded | 69802fdda8b01e76cc8e185a |
| LLaVA-Med-7B | 16.4 | Not recorded | 69802fdda8b01e76cc8e185a |
| GPT-5 | 14.7 | Not recorded | 69802fdda8b01e76cc8e185a |

### Lingshu-7B + Med-Scout

ROUGE-L: 31.4. Result ID: 69976ed45698665ec4c2af17.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs","Post-training strategy":"Med-Scout"}. This record does not assert independent reproduction.

### Lingshu-7B

ROUGE-L: 30.9. Result ID: 69976ed45698665ec4c2af16.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795a1e209131cea7d387.md>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs"}. This record does not assert independent reproduction.

### MedGemma-4B-IT

ROUGE-L: 27.4. Result ID: 69976ed45698665ec4c2af13.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ff1e209131cea808a4.md>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs"}. This record does not assert independent reproduction.

### HuatuoGPT-Vision-7B + Med-Scout

ROUGE-L: 25.7. Result ID: 69976ed45698665ec4c2af15.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs","Post-training strategy":"Med-Scout"}. This record does not assert independent reproduction.

### GEMINI-3-FLASH

ROUGE-L: 24.9. Result ID: 69976ed45698665ec4c2af0a.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

### Qwen3-VL-8B-Instruct + Med-Scout

ROUGE-L: 23.8. Result ID: 69976ed45698665ec4c2af11.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs","Post-training strategy":"Med-Scout"}. This record does not assert independent reproduction.

### HuatuoGPT-Vision-7B

ROUGE-L: 23.6. Result ID: 69976ed45698665ec4c2af14.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f71e209131cea64ea0.md>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs"}. This record does not assert independent reproduction.

### Qwen3-VL-4B-Instruct + Med-Scout

ROUGE-L: 23.4. Result ID: 69976ed45698665ec4c2af0f.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs","Post-training strategy":"Med-Scout"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B-Instruct

ROUGE-L: 22.9. Result ID: 69976ed45698665ec4c2af0d.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs"}. This record does not assert independent reproduction.

### Qwen3-VL-4B-Instruct

ROUGE-L: 21.8. Result ID: 69976ed45698665ec4c2af0e.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs"}. This record does not assert independent reproduction.

### Qwen2.5-VL-3B-Instruct

ROUGE-L: 21.4. Result ID: 69976ed45698665ec4c2af0c.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs"}. This record does not assert independent reproduction.

### Qwen3-VL-8B-Instruct

ROUGE-L: 21.3. Result ID: 69976ed45698665ec4c2af10.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs"}. This record does not assert independent reproduction.

### InternVL3-8B

ROUGE-L: 20.7. Result ID: 69976ed45698665ec4c2af0b.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76fa5.md>)

Recorded evaluation configuration: {"Model Category":"General-purpose MLLMs"}. This record does not assert independent reproduction.

### LLaVA-Med-7B

ROUGE-L: 16.4. Result ID: 69976ed45698665ec4c2af12.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea46356.md>)

[Implementation](<https://aka.ms/llava-med>)

Recorded evaluation configuration: {"Model Category":"Medical MLLMs"}. This record does not assert independent reproduction.

### GPT-5

ROUGE-L: 14.7. Result ID: 69976ed45698665ec4c2af09.

[Reporting paper](<https://www.sota2.com/research/paper/69802fdda8b01e76cc8e185a.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

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