# Cognitive Distraction Detection on DR(eye)VE (Leave-one-dataset-out)

Source: [Cognitive Distraction Detection on DR(eye)VE (Leave-one-dataset-out)](<https://www.sota2.com/research/sota/cognitive-distraction-detection-on-dr-eye-ve-leave-one-dataset-out>)

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

Data updated: 2026-05-11T10:30:21.639Z

Evaluation of Cognitive Distraction Detection performance on the DR(eye)VE dataset using a Leave-one-dataset-out cross-validation setup, 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 | Cognitive Distraction Detection |
| Dataset (including any stated split) | DR(eye)VE (Leave-one-dataset-out) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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/cognitive-distraction-detection-on-dr-eye-ve-leave-one-dataset-out.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/cognitive-distraction-detection-on-dr-eye-ve-leave-one-dataset-out.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| EyeCue | 60.2 | Not recorded | 6a0162c0dbc836b87d8233aa |
| InternVideo2s1-1B | 59.31 | Not recorded | 6a0162c0dbc836b87d8233aa |
| VideoMAEK400 | 58.85 | Not recorded | 6a0162c0dbc836b87d8233aa |
| EgoVideo | 55.65 | Not recorded | 6a0162c0dbc836b87d8233aa |
| TimeSformerK600 | 55.41 | Not recorded | 6a0162c0dbc836b87d8233aa |
| TimeSformerK400 | 55.4 | Not recorded | 6a0162c0dbc836b87d8233aa |
| Voila-A | 54.91 | Not recorded | 6a0162c0dbc836b87d8233aa |
| EyeCue | 54.67 | Not recorded | 6a0162c0dbc836b87d8233aa |
| Video-LLaVA | 53.75 | Not recorded | 6a0162c0dbc836b87d8233aa |
| GazeVQA | 51.97 | Not recorded | 6a0162c0dbc836b87d8233aa |
| GazeGPT | 51.29 | Not recorded | 6a0162c0dbc836b87d8233aa |
| VideoLLaMA3 | 51.17 | Not recorded | 6a0162c0dbc836b87d8233aa |
| Heatmap-based | 49.39 | Not recorded | 6a0162c0dbc836b87d8233aa |
| GazeLLM | 49.14 | Not recorded | 6a0162c0dbc836b87d8233aa |

### EyeCue

Accuracy: 60.2. Result ID: 6a01afba47f34ce924e43e4c.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

Recorded evaluation configuration: {"Backbone":"TimeSformerK600","Training-free":"false"}. This record does not assert independent reproduction.

### InternVideo2s1-1B

Accuracy: 59.31. Result ID: 6a01afba47f34ce924e43e44.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74871e209131cea623f3.md>)

[Implementation](<https://github.com/OpenGVLab/InternVideo2>)

Recorded evaluation configuration: {"Backbone":"ViT-B/14","Training-free":"false"}. This record does not assert independent reproduction.

### VideoMAEK400

Accuracy: 58.85. Result ID: 6a01afba47f34ce924e43e42.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da81e209131cea3b2fd.md>)

[Implementation](<https://github.com/MCG-NJU/VideoMAE>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Training-free":"false"}. This record does not assert independent reproduction.

### EgoVideo

Accuracy: 55.65. Result ID: 6a01afba47f34ce924e43e43.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ea1e209131cea5ee05.md>)

[Implementation](<https://github.com/OpenGVLab/EgoVideo>)

Recorded evaluation configuration: {"Backbone":"ViT-B/14","Training-free":"false"}. This record does not assert independent reproduction.

### TimeSformerK600

Accuracy: 55.41. Result ID: 6a01afba47f34ce924e43e41.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a1d1e209131cea25fbb.md>)

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Training-free":"false"}. This record does not assert independent reproduction.

### TimeSformerK400

Accuracy: 55.4. Result ID: 6a01afba47f34ce924e43e40.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a1d1e209131cea25fbb.md>)

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Training-free":"false"}. This record does not assert independent reproduction.

### Voila-A

Accuracy: 54.91. Result ID: 6a01afba47f34ce924e43e48.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71e11e209131cea53885.md>)

Recorded evaluation configuration: {"Backbone":"CLIP ViT-L/14","Training-free":"false"}. This record does not assert independent reproduction.

### EyeCue

Accuracy: 54.67. Result ID: 6a01afba47f34ce924e43e4b.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

Recorded evaluation configuration: {"Backbone":"VideoMAEK400","Training-free":"false"}. This record does not assert independent reproduction.

### Video-LLaVA

Accuracy: 53.75. Result ID: 6a01afba47f34ce924e43e45.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65534.md>)

[Implementation](<https://github.com/PKU-YuanGroup/Video-LLaVA>)

Recorded evaluation configuration: {"Backbone":"OpenCLIP-L/14","Training-free":"false"}. This record does not assert independent reproduction.

### GazeVQA

Accuracy: 51.97. Result ID: 6a01afba47f34ce924e43e49.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

Recorded evaluation configuration: {"Backbone":"CLIP ViT-B/32","Training-free":"false"}. This record does not assert independent reproduction.

### GazeGPT

Accuracy: 51.29. Result ID: 6a01afba47f34ce924e43e47.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71fd1e209131cea53f25.md>)

Recorded evaluation configuration: {"Backbone":"GLM-4.5V-AWQ","Training-free":"true"}. This record does not assert independent reproduction.

### VideoLLaMA3

Accuracy: 51.17. Result ID: 6a01afba47f34ce924e43e46.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79111e209131cea7bf26.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Training-free":"false"}. This record does not assert independent reproduction.

### Heatmap-based

Accuracy: 49.39. Result ID: 6a01afba47f34ce924e43e3f.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f151e209131ce9e75c8.md>)

Recorded evaluation configuration: {"Backbone":"SVM","Training-free":"false"}. This record does not assert independent reproduction.

### GazeLLM

Accuracy: 49.14. Result ID: 6a01afba47f34ce924e43e4a.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162c0dbc836b87d8233aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77d91e209131cea7587c.md>)

Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Pro","Training-free":"true"}. This record does not assert independent reproduction.

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