# Forensic Attribution on FakeBench MCQ

Source: [Forensic Attribution on FakeBench MCQ](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq>)

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

Data updated: 2026-03-17T07:00:51.179Z

Evaluation of Forensic Attribution performance on the FakeBench MCQ dataset, measured across various qualitative and overall metrics.

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 | Forensic Attribution |
| Dataset (including any stated split) | FakeBench MCQ |
| Metric | Yes/No Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [Yes/No Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=Yes%2FNo%20Accuracy>): Higher is better; unit: PERCENT.

- [What Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=What%20Accuracy>): Higher is better; unit: PERCENT.

- [Where Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=Where%20Accuracy>): Higher is better; unit: PERCENT.

- [How Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=How%20Accuracy>): Higher is better; unit: PERCENT.

- [Low Difficulty Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=Low%20Difficulty%20Accuracy>): Higher is better; unit: PERCENT.

- [Medium Difficulty Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=Medium%20Difficulty%20Accuracy>): Higher is better; unit: PERCENT.

- [High Difficulty Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=High%20Difficulty%20Accuracy>): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/forensic-attribution-on-fakebench-mcq.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Yes/No Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4V (Teacher) | 79.47 | Not recorded | 68fa77af1e209131cea74b68 |
| FakeScope-L | 79.47 | Not recorded | 68fa77af1e209131cea74b68 |
| FakeScope-M | 75.47 | Not recorded | 68fa77af1e209131cea74b68 |
| mPLUG-Owl2 | 70.53 | Not recorded | 68fa77af1e209131cea74b68 |
| GeminiPro | 70.37 | Not recorded | 68fa77af1e209131cea74b68 |
| InternLM-XC.2-vl | 69.73 | Not recorded | 68fa77af1e209131cea74b68 |
| Claude3 Sonnet | 68.27 | Not recorded | 68fa77af1e209131cea74b68 |
| Qwen-VL | 64.13 | Not recorded | 68fa77af1e209131cea74b68 |
| LLaVA-v1.5 | 60.4 | Not recorded | 68fa77af1e209131cea74b68 |
| InstructBLIP | 56.27 | Not recorded | 68fa77af1e209131cea74b68 |
| Random guess | 50.13 | Not recorded | 68fa77af1e209131cea74b68 |
| IDEFICS-Instruct | 48.67 | Not recorded | 68fa77af1e209131cea74b68 |

### GPT-4V (Teacher)

Yes/No Accuracy: 79.47. Result ID: 69b8fc1b47f34ce924a0ed52.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### FakeScope-L

Yes/No Accuracy: 79.47. Result ID: 69b8fc1b47f34ce924a0ed5c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

Recorded evaluation configuration: {"Base Model":"LLaVA-1.5"}. This record does not assert independent reproduction.

### FakeScope-M

Yes/No Accuracy: 75.47. Result ID: 69b8fc1b47f34ce924a0ed5b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

Recorded evaluation configuration: {"Base Model":"mPLUG-Owl2"}. This record does not assert independent reproduction.

### mPLUG-Owl2

Yes/No Accuracy: 70.53. Result ID: 69b8fc1b47f34ce924a0ed5a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71201e209131cea4f575.md>)

[Implementation](<https://github.com/X-PLUG/mPLUG-Owl/tree/main/mPLUG-Owl2>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GeminiPro

Yes/No Accuracy: 70.37. Result ID: 69b8fc1b47f34ce924a0ed53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78731e209131cea78622.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### InternLM-XC.2-vl

Yes/No Accuracy: 69.73. Result ID: 69b8fc1b47f34ce924a0ed57.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72021e209131cea541e3.md>)

[Implementation](<https://github.com/InternLM/InternLM-XComposer>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Claude3 Sonnet

Yes/No Accuracy: 68.27. Result ID: 69b8fc1b47f34ce924a0ed54.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen-VL

Yes/No Accuracy: 64.13. Result ID: 69b8fc1b47f34ce924a0ed59.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

[Implementation](<https://github.com/QwenLM/Qwen-VL>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LLaVA-v1.5

Yes/No Accuracy: 60.4. Result ID: 69b8fc1b47f34ce924a0ed58.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### InstructBLIP

Yes/No Accuracy: 56.27. Result ID: 69b8fc1b47f34ce924a0ed55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea47021.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/instructblip>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Random guess

Yes/No Accuracy: 50.13. Result ID: 69b8fc1b47f34ce924a0ed51.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### IDEFICS-Instruct

Yes/No Accuracy: 48.67. Result ID: 69b8fc1b47f34ce924a0ed56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70581e209131cea4a712.md>)

[Implementation](<https://github.com/huggingface/OBELICS>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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