# Forgery Detection on MedForge-90K Implant Forgery Cross-Model

Source: [Forgery Detection on MedForge-90K Implant Forgery Cross-Model](<https://www.sota2.com/research/sota/forgery-detection-on-medforge-90k-implant-forgery-cross-model>)

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

Data updated: 2026-03-20T09:21:56.555Z

Evaluation of Forgery Detection performance on the MedForge-90K Implant Forgery Cross-Model dataset, reporting Accuracy and F1 Score.

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 | Forgery Detection |
| Dataset (including any stated split) | MedForge-90K Implant Forgery Cross-Model |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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/forgery-detection-on-medforge-90k-implant-forgery-cross-model.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/forgery-detection-on-medforge-90k-implant-forgery-cross-model.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MedForge-Reasoner | 94.86 | Not recorded | 69bcd4770636e5bf4afc297d |
| SIDA-13B | 85.86 | Not recorded | 69bcd4770636e5bf4afc297d |
| AIGC-Holmes | 80.91 | Not recorded | 69bcd4770636e5bf4afc297d |
| FakeVLM | 79.82 | Not recorded | 69bcd4770636e5bf4afc297d |
| SIDA-7B | 76.57 | Not recorded | 69bcd4770636e5bf4afc297d |
| Gemini3-Pro | 74.8 | Not recorded | 69bcd4770636e5bf4afc297d |
| Gemini3-Flash | 60.14 | Not recorded | 69bcd4770636e5bf4afc297d |
| Qwen3VL-Plus | 56.03 | Not recorded | 69bcd4770636e5bf4afc297d |
| Qwen3VL-Flash | 50.93 | Not recorded | 69bcd4770636e5bf4afc297d |

### MedForge-Reasoner

Accuracy: 94.86. Result ID: 69bd11a247f34ce924a763aa.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

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

### SIDA-13B

Accuracy: 85.86. Result ID: 69bd11a247f34ce924a763a3.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a11e209131cea7e8bc.md>)

[Implementation](<https://hzlsaber.github.io/projects/SIDA/>)

Recorded evaluation configuration: {"Category":"Specialized Detectors"}. This record does not assert independent reproduction.

### AIGC-Holmes

Accuracy: 80.91. Result ID: 69bd11a247f34ce924a763a5.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79db1e209131cea7fc5a.md>)

Recorded evaluation configuration: {"Category":"Specialized Detectors"}. This record does not assert independent reproduction.

### FakeVLM

Accuracy: 79.82. Result ID: 69bd11a247f34ce924a763a4.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c301e209131cea8ce96.md>)

Recorded evaluation configuration: {"Category":"Specialized Detectors"}. This record does not assert independent reproduction.

### SIDA-7B

Accuracy: 76.57. Result ID: 69bd11a247f34ce924a763a2.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a11e209131cea7e8bc.md>)

[Implementation](<https://hzlsaber.github.io/projects/SIDA/>)

Recorded evaluation configuration: {"Category":"Specialized Detectors"}. This record does not assert independent reproduction.

### Gemini3-Pro

Accuracy: 74.8. Result ID: 69bd11a247f34ce924a763a9.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

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

### Gemini3-Flash

Accuracy: 60.14. Result ID: 69bd11a247f34ce924a763a8.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

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

### Qwen3VL-Plus

Accuracy: 56.03. Result ID: 69bd11a247f34ce924a763a7.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

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

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

### Qwen3VL-Flash

Accuracy: 50.93. Result ID: 69bd11a247f34ce924a763a6.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd4770636e5bf4afc297d.md>)

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

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

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