# Biomedical Question Answering on PubMedQA PQA-L In-Domain (test)

Source: [Biomedical Question Answering on PubMedQA PQA-L In-Domain (test)](<https://www.sota2.com/research/sota/biomedical-question-answering-on-pubmedqa-pqa-l-in-domain-test>)

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

Data updated: 2026-02-26T05:29:04.636Z

Evaluation of model performance on the In-Domain test split of the PQA-L subset of the PubMedQA dataset for Biomedical Question Answering.

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 | Biomedical Question Answering |
| Dataset (including any stated split) | PubMedQA PQA-L In-Domain (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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/biomedical-question-answering-on-pubmedqa-pqa-l-in-domain-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Human (expert) | 78 | Not recorded | 68fa70591e209131cea4a74b |
| BioMedGPT-10B | 76.1 | Not recorded | 68fa70591e209131cea4a74b |
| Llama2-Chat | 75.5 | Not recorded | 68fa70591e209131cea4a74b |
| Llama | 73.4 | Not recorded | 68fa70591e209131cea4a74b |
| InstructGPT | 73.2 | Not recorded | 68fa70591e209131cea4a74b |
| PMC-Llama | 69.5 | Not recorded | 68fa70591e209131cea4a74b |
| ChatGPT | 63.9 | Not recorded | 68fa70591e209131cea4a74b |
| Human (pass) | 60 | Not recorded | 68fa70591e209131cea4a74b |
| Llama2-Chat | 21.9 | Not recorded | 68fa70591e209131cea4a74b |
| Llama | 5.2 | Not recorded | 68fa70591e209131cea4a74b |
| Llama2 | 3.7 | Not recorded | 68fa70591e209131cea4a74b |

### Human (expert)

Accuracy: 78. Result ID: 69912ace3a6c314849062f8a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa731a1e209131cea5a8e6.md>)

Recorded evaluation configuration: {"Setting":"Manual"}. This record does not assert independent reproduction.

### BioMedGPT-10B

Accuracy: 76.1. Result ID: 69912ace3a6c314849062f93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Implementation](<https://github.com/PharMolix/OpenBioMed>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### Llama2-Chat

Accuracy: 75.5. Result ID: 69912ace3a6c314849062f91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701f1e209131cea492fe.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### Llama

Accuracy: 73.4. Result ID: 69912ace3a6c314849062f90.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eae1e209131cea41685.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### InstructGPT

Accuracy: 73.2. Result ID: 69912ace3a6c314849062f8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf71e209131cea314c0.md>)

[Implementation](<https://github.com/openai/following-instructions-human-feedback>)

Recorded evaluation configuration: {"Setting":"Zero-shot"}. This record does not assert independent reproduction.

### PMC-Llama

Accuracy: 69.5. Result ID: 69912ace3a6c314849062f92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70601e209131cea4ab5d.md>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### ChatGPT

Accuracy: 63.9. Result ID: 69912ace3a6c314849062f8c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

Recorded evaluation configuration: {"Setting":"Zero-shot"}. This record does not assert independent reproduction.

### Human (pass)

Accuracy: 60. Result ID: 69912ace3a6c314849062f89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa731a1e209131cea5a8e6.md>)

Recorded evaluation configuration: {"Setting":"Manual"}. This record does not assert independent reproduction.

### Llama2-Chat

Accuracy: 21.9. Result ID: 69912ace3a6c314849062f8f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701f1e209131cea492fe.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Setting":"Zero-shot"}. This record does not assert independent reproduction.

### Llama

Accuracy: 5.2. Result ID: 69912ace3a6c314849062f8d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eae1e209131cea41685.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Setting":"Zero-shot"}. This record does not assert independent reproduction.

### Llama2

Accuracy: 3.7. Result ID: 69912ace3a6c314849062f8e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a74b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701f1e209131cea492fe.md>)

[Implementation](<https://github.com/facebookresearch/llama>)

Recorded evaluation configuration: {"Setting":"Zero-shot"}. This record does not assert independent reproduction.

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