# Medical Question Answering on MedQA USMLE (τ, Speedup)

Source: [Medical Question Answering on MedQA USMLE (τ, Speedup)](<https://www.sota2.com/research/sota/medical-question-answering-on-medqa-usmle-speedup>)

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

Data updated: 2026-03-11T08:59:57.613Z

Evaluation of model performance on the MedQA USMLE dataset, reporting Kendall's Tau (τ) and inference Speedup.

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 Question Answering |
| Dataset (including any stated split) | MedQA USMLE |
| Metric | Kendall's Tau (τ) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Kendall's Tau (τ)](<https://www.sota2.com/research/sota/medical-question-answering-on-medqa-usmle-speedup.md?metric=Kendall%27s%20Tau%20(%CF%84)>): Higher is better; unit: SCALAR.

- [Speedup](<https://www.sota2.com/research/sota/medical-question-answering-on-medqa-usmle-speedup.md?metric=Speedup>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Kendall's Tau (τ) | Result date | Report paper |
| --- | --- | --- | --- |
| EDA (Ours) | 4.34 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| Full-FT | 3.97 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| EDA (Base) | 3.94 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| LoRA | 3.9 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| EDA (Ours) | 3.12 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| Full-FT | 2.65 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| EDA (Base) | 2.62 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| LoRA | 2.6 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| Training-Free | 2.17 | Not recorded | 69b0f72d2d603e6de3e45f9c |
| Training-Free | 1.55 | Not recorded | 69b0f72d2d603e6de3e45f9c |

### EDA (Ours)

Kendall's Tau (τ): 4.34. Result ID: 69b12f076dbb7075b95bbbd9.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"0","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### Full-FT

Kendall's Tau (τ): 3.97. Result ID: 69b12f076dbb7075b95bbbd6.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"0","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### EDA (Base)

Kendall's Tau (τ): 3.94. Result ID: 69b12f076dbb7075b95bbbd8.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"0","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### LoRA

Kendall's Tau (τ): 3.9. Result ID: 69b12f076dbb7075b95bbbd7.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"Temperature":"0","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### EDA (Ours)

Kendall's Tau (τ): 3.12. Result ID: 69b12f076dbb7075b95bbbde.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"1","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### Full-FT

Kendall's Tau (τ): 2.65. Result ID: 69b12f076dbb7075b95bbbdb.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"1","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### EDA (Base)

Kendall's Tau (τ): 2.62. Result ID: 69b12f076dbb7075b95bbbdd.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"1","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### LoRA

Kendall's Tau (τ): 2.6. Result ID: 69b12f076dbb7075b95bbbdc.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"Temperature":"1","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### Training-Free

Kendall's Tau (τ): 2.17. Result ID: 69b12f076dbb7075b95bbbd5.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"0","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

### Training-Free

Kendall's Tau (τ): 1.55. Result ID: 69b12f076dbb7075b95bbbda.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f72d2d603e6de3e45f9c.md>)

Recorded evaluation configuration: {"Temperature":"1","Target Model":"Meditron3-Qwen2.5-7B"}. This record does not assert independent reproduction.

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