# Video Robustness Classification on ImageNet-Vid-Robust (test)

Source: [Video Robustness Classification on ImageNet-Vid-Robust (test)](<https://www.sota2.com/research/sota/video-robustness-classification-on-imagenet-vid-robust-test>)

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

Data updated: 2026-02-26T05:34:22.400Z

Evaluation of model performance on the ImageNet-Vid-Robust test set, focusing on video classification robustness by comparing original accuracy against accuracy under perturbation.

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 | Video Robustness Classification |
| Dataset (including any stated split) | ImageNet-Vid-Robust (test) |
| Metric | Accuracy (Original) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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 (Original)](<https://www.sota2.com/research/sota/video-robustness-classification-on-imagenet-vid-robust-test.md?metric=Accuracy%20(Original)>): Higher is better; unit: PERCENT.

- [Accuracy (Perturbed)](<https://www.sota2.com/research/sota/video-robustness-classification-on-imagenet-vid-robust-test.md?metric=Accuracy%20(Perturbed)>): Higher is better; unit: PERCENT.

- [Accuracy Drop](<https://www.sota2.com/research/sota/video-robustness-classification-on-imagenet-vid-robust-test.md?metric=Accuracy%20Drop>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Original) | Result date | Report paper |
| --- | --- | --- | --- |
| ImageNet (Supervised) | 68 | Not recorded | 68fa67561e209131cea159ce |
| VIVI-Ex(4)-Co(100%) | 62.2 | Not recorded | 68fa67561e209131cea159ce |

### ImageNet (Supervised)

Accuracy (Original): 68. Result ID: 698d2e2b8b517eb4331f986b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159ce.md>)

Recorded evaluation configuration: {"Training Mode":"Purely Supervised"}. This record does not assert independent reproduction.

### VIVI-Ex(4)-Co(100%)

Accuracy (Original): 62.2. Result ID: 698d2e2b8b517eb4331f986c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159ce.md>)

[Implementation](<https://tfhub.dev/google/image_augmentation/nas_imagenet/1>)

Recorded evaluation configuration: {"Training Mode":"Co-trained","ImageNet Labels Used":"100%","Base Model":"Exemplar-based (Ex(4))"}. This record does not assert independent reproduction.

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