# Medical Multi-choice Question Answering on MMedBench (test)

Source: [Medical Multi-choice Question Answering on MMedBench (test)](<https://www.sota2.com/research/sota/medical-multi-choice-question-answering-on-mmedbench-test>)

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

Data updated: 2026-02-26T06:06:34.094Z

Evaluation of model performance on the test split of the MMedBench dataset for Medical Multi-choice Question Answering, measured using Token Perplexity.

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 Multi-choice Question Answering |
| Dataset (including any stated split) | MMedBench (test) |
| Metric | Token Perplexity (log) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 16 of 16 |
| 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

- [Token Perplexity (log)](<https://www.sota2.com/research/sota/medical-multi-choice-question-answering-on-mmedbench-test.md?metric=Token%20Perplexity%20(log)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Token Perplexity (log) | Result date | Report paper |
| --- | --- | --- | --- |
| OFA-KD | 0.1494 | Not recorded | 6993f6419ebc499c97409c62 |
| DEEPFUSION | 0.1494 | Not recorded | 6993f6419ebc499c97409c62 |
| OFA-KD | 0.151 | Not recorded | 6993f6419ebc499c97409c62 |
| FedKMT | 0.1526 | Not recorded | 6993f6419ebc499c97409c62 |
| OFA-KD | 0.1535 | Not recorded | 6993f6419ebc499c97409c62 |
| FedKMT | 0.1546 | Not recorded | 6993f6419ebc499c97409c62 |
| DEEPFUSION | 0.1551 | Not recorded | 6993f6419ebc499c97409c62 |
| FedJETS | 0.1557 | Not recorded | 6993f6419ebc499c97409c62 |
| DEEPFUSION | 0.1569 | Not recorded | 6993f6419ebc499c97409c62 |
| FedJETS | 0.1571 | Not recorded | 6993f6419ebc499c97409c62 |
| OFA-KD | 0.1573 | Not recorded | 6993f6419ebc499c97409c62 |
| FedKMT | 0.1574 | Not recorded | 6993f6419ebc499c97409c62 |
| FedJETS | 0.1588 | Not recorded | 6993f6419ebc499c97409c62 |
| DEEPFUSION | 0.1603 | Not recorded | 6993f6419ebc499c97409c62 |
| FedKMT | 0.1609 | Not recorded | 6993f6419ebc499c97409c62 |
| FedJETS | 0.1639 | Not recorded | 6993f6419ebc499c97409c62 |

### OFA-KD

Token Perplexity (log): 0.1494. Result ID: 699a96955698665ec421b8f2.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71061e209131cea4ea5e.md>)

[Implementation](<https://github.com/Hao840/OFAKD>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"128"}. This record does not assert independent reproduction.

### DEEPFUSION

Token Perplexity (log): 0.1494. Result ID: 699a96955698665ec421b8f6.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"128"}. This record does not assert independent reproduction.

### OFA-KD

Token Perplexity (log): 0.151. Result ID: 699a96955698665ec421b8f1.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71061e209131cea4ea5e.md>)

[Implementation](<https://github.com/Hao840/OFAKD>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"64"}. This record does not assert independent reproduction.

### FedKMT

Token Perplexity (log): 0.1526. Result ID: 699a96955698665ec421b8ed.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76371e209131cea6be4c.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"64"}. This record does not assert independent reproduction.

### OFA-KD

Token Perplexity (log): 0.1535. Result ID: 699a96955698665ec421b8f0.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71061e209131cea4ea5e.md>)

[Implementation](<https://github.com/Hao840/OFAKD>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"32"}. This record does not assert independent reproduction.

### FedKMT

Token Perplexity (log): 0.1546. Result ID: 699a96955698665ec421b8ee.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76371e209131cea6be4c.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"128"}. This record does not assert independent reproduction.

### DEEPFUSION

Token Perplexity (log): 0.1551. Result ID: 699a96955698665ec421b8f5.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"64"}. This record does not assert independent reproduction.

### FedJETS

Token Perplexity (log): 0.1557. Result ID: 699a96955698665ec421b8e9.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"64"}. This record does not assert independent reproduction.

### DEEPFUSION

Token Perplexity (log): 0.1569. Result ID: 699a96955698665ec421b8f4.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"32"}. This record does not assert independent reproduction.

### FedJETS

Token Perplexity (log): 0.1571. Result ID: 699a96955698665ec421b8ea.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"128"}. This record does not assert independent reproduction.

### OFA-KD

Token Perplexity (log): 0.1573. Result ID: 699a96955698665ec421b8ef.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71061e209131cea4ea5e.md>)

[Implementation](<https://github.com/Hao840/OFAKD>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"16"}. This record does not assert independent reproduction.

### FedKMT

Token Perplexity (log): 0.1574. Result ID: 699a96955698665ec421b8ec.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76371e209131cea6be4c.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"32"}. This record does not assert independent reproduction.

### FedJETS

Token Perplexity (log): 0.1588. Result ID: 699a96955698665ec421b8e8.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"32"}. This record does not assert independent reproduction.

### DEEPFUSION

Token Perplexity (log): 0.1603. Result ID: 699a96955698665ec421b8f3.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"16"}. This record does not assert independent reproduction.

### FedKMT

Token Perplexity (log): 0.1609. Result ID: 699a96955698665ec421b8eb.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76371e209131cea6be4c.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"16"}. This record does not assert independent reproduction.

### FedJETS

Token Perplexity (log): 0.1639. Result ID: 699a96955698665ec421b8e7.

[Reporting paper](<https://www.sota2.com/research/paper/6993f6419ebc499c97409c62.md>)

Recorded evaluation configuration: {"Number of participating edge devices (N)":"16"}. This record does not assert independent reproduction.

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