# Video Reasoning on WorldSense

Source: [Video Reasoning on WorldSense](<https://www.sota2.com/research/sota/video-reasoning-on-worldsense>)

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

Data updated: 2026-04-14T12:24:18.363Z

Evaluation of model performance on the WorldSense dataset for Video Reasoning tasks, measured by Accuracy and Avg.

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 Reasoning |
| Dataset (including any stated split) | WorldSense |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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/video-reasoning-on-worldsense.md?metric=Accuracy>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| POINTS1.5-8B (baseline) | 40.4 | Not recorded | 69ddca7d7d0172140ec79f9a |
| POINTS-Long (focus) | 40.4 | Not recorded | 69ddca7d7d0172140ec79f9a |
| LLaVA-Video-7B | 40.2 | Not recorded | 69ddca7d7d0172140ec79f9a |
| POINTS-Long (standby) | 40 | Not recorded | 69ddca7d7d0172140ec79f9a |
| POINTS-Long (standby) | 39.7 | Not recorded | 69ddca7d7d0172140ec79f9a |
| POINTS-Long (standby) | 39.4 | Not recorded | 69ddca7d7d0172140ec79f9a |
| LLaVA-OV-7B | 37.7 | Not recorded | 69ddca7d7d0172140ec79f9a |

### POINTS1.5-8B (baseline)

Accuracy: 40.4. Result ID: 69de31e047f34ce924c702fa.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76211e209131cea6b618.md>)

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"324","Total Num of Token":"≈ 20K"}. This record does not assert independent reproduction.

### POINTS-Long (focus)

Accuracy: 40.4. Result ID: 69de31e047f34ce924c702fe.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"324+32","Total Num of Token":"≈ 22K"}. This record does not assert independent reproduction.

### LLaVA-Video-7B

Accuracy: 40.2. Result ID: 69de31e047f34ce924c702f9.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a561e209131cea82934.md>)

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

Recorded evaluation configuration: {"Num Frame":"-","Token/Frame":"-","Total Num of Token":"\>12k"}. This record does not assert independent reproduction.

### POINTS-Long (standby)

Accuracy: 40. Result ID: 69de31e047f34ce924c702fd.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"32","Total Num of Token":"2048 (10%)"}. This record does not assert independent reproduction.

### POINTS-Long (standby)

Accuracy: 39.7. Result ID: 69de31e047f34ce924c702fc.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"16","Total Num of Token":"1024 (5%)"}. This record does not assert independent reproduction.

### POINTS-Long (standby)

Accuracy: 39.4. Result ID: 69de31e047f34ce924c702fb.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"8","Total Num of Token":"512 (2.5%)"}. This record does not assert independent reproduction.

### LLaVA-OV-7B

Accuracy: 37.7. Result ID: 69de31e047f34ce924c702f8.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e58.md>)

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

Recorded evaluation configuration: {"Num Frame":"64","Token/Frame":"-","Total Num of Token":"\>12k"}. This record does not assert independent reproduction.

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