# Video Perception on WorldArena

Source: [Video Perception on WorldArena](<https://www.sota2.com/research/sota/video-perception-on-worldarena>)

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

Data updated: 2026-05-08T10:22:35.720Z

Evaluation of model performance on the WorldArena dataset for Video Perception tasks, covering metrics like JEPA, Flow, Depth, and Action scores.

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 Perception |
| Dataset (including any stated split) | WorldArena |
| Metric | Img Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Img Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Img%20Score>): Higher is better; unit: SCALAR.

- [Aesthetic Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Aesthetic%20Score>): Higher is better; unit: SCALAR.

- [JEPA Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=JEPA%20Score>): Higher is better; unit: SCALAR.

- [Dynamics Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Dynamics%20Score>): Higher is better; unit: SCALAR.

- [Optical Flow Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Optical%20Flow%20Score>): Higher is better; unit: SCALAR.

- [Smoothness Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Smoothness%20Score>): Higher is better; unit: SCALAR.

- [Subjective Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Subjective%20Score>): Higher is better; unit: SCALAR.

- [Background Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Background%20Score>): Higher is better; unit: SCALAR.

- [Photorealism Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Photorealism%20Score>): Higher is better; unit: SCALAR.

- [Interaction Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Interaction%20Score>): Higher is better; unit: SCALAR.

- [Trajectory Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Trajectory%20Score>): Higher is better; unit: SCALAR.

- [Depth Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Depth%20Score>): Higher is better; unit: SCALAR.

- [Perspective Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Perspective%20Score>): Higher is better; unit: SCALAR.

- [Instruction Following Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Instruction%20Following%20Score>): Higher is better; unit: SCALAR.

- [Semantic Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Semantic%20Score>): Higher is better; unit: SCALAR.

- [Action Score](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Action%20Score>): Higher is better; unit: SCALAR.

- [Average Score (Avg16)](<https://www.sota2.com/research/sota/video-perception-on-worldarena.md?metric=Average%20Score%20(Avg16)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Img Score | Result date | Report paper |
| --- | --- | --- | --- |
| Cosmos-Predict 2.5 (action) | 0.449 | Not recorded | 69fd6e1b5497b3b670c89471 |
| KVAF-conditioned generation | 0.364 | Not recorded | 69fd6e1b5497b3b670c89471 |
| CtrlWorld | 0.352 | Not recorded | 69fd6e1b5497b3b670c89471 |
| IRASim | 0.349 | Not recorded | 69fd6e1b5497b3b670c89471 |
| RoboMaster | 0.349 | Not recorded | 69fd6e1b5497b3b670c89471 |

### Cosmos-Predict 2.5 (action)

Img Score: 0.449. Result ID: 69fdb96447f34ce924e14ea3.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

Recorded evaluation configuration: {"Conditioning":"Action"}. This record does not assert independent reproduction.

### KVAF-conditioned generation

Img Score: 0.364. Result ID: 69fdb96447f34ce924e14ea5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

Recorded evaluation configuration: {"Conditioning":"KVAF"}. This record does not assert independent reproduction.

### CtrlWorld

Img Score: 0.352. Result ID: 69fdb96447f34ce924e14ea1.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

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

### IRASim

Img Score: 0.349. Result ID: 69fdb96447f34ce924e14ea2.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4c1e209131cea8247a.md>)

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

### RoboMaster

Img Score: 0.349. Result ID: 69fdb96447f34ce924e14ea4.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1b5497b3b670c89471.md>)

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.
