# Video Reasoning on SEED-Bench L3 OOD R1

Source: [Video Reasoning on SEED-Bench L3 OOD R1](<https://www.sota2.com/research/sota/video-reasoning-on-seed-bench-l3-ood-r1>)

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

Data updated: 2026-03-04T18:46:50.815Z

Evaluation of Video Reasoning capabilities using the SEED-Bench L3 OOD R1 subset, measured by Accuracy.

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) | SEED-Bench L3 OOD R1 |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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-seed-bench-l3-ood-r1.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| APPO | 49.3 | Not recorded | 69a5195f2d603e6de3e3c777 |
| DAPO | 48.7 | Not recorded | 69a5195f2d603e6de3e3c777 |
| GRPO | 48.2 | Not recorded | 69a5195f2d603e6de3e3c777 |
| SFT | 43.9 | Not recorded | 69a5195f2d603e6de3e3c777 |
| APPO | 35 | Not recorded | 69a5195f2d603e6de3e3c777 |
| APPO | 34.7 | Not recorded | 69a5195f2d603e6de3e3c777 |
| DAPO | 31.8 | Not recorded | 69a5195f2d603e6de3e3c777 |
| GRPO | 31 | Not recorded | 69a5195f2d603e6de3e3c777 |
| Base Model | 29.6 | Not recorded | 69a5195f2d603e6de3e3c777 |
| Video-R1 | 28.8 | Not recorded | 69a5195f2d603e6de3e3c777 |
| SFT | 28.6 | Not recorded | 69a5195f2d603e6de3e3c777 |
| VideoChat-R1 | 27.8 | Not recorded | 69a5195f2d603e6de3e3c777 |
| VideoRFT | 27.7 | Not recorded | 69a5195f2d603e6de3e3c777 |
| Base Model | 27 | Not recorded | 69a5195f2d603e6de3e3c777 |
| GRPO-CARE | 26 | Not recorded | 69a5195f2d603e6de3e3c777 |
| TW-GRPO | 25.6 | Not recorded | 69a5195f2d603e6de3e3c777 |

### APPO

Accuracy: 49.3. Result ID: 69a87e046dbb7075b950bb44.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### DAPO

Accuracy: 48.7. Result ID: 69a87e046dbb7075b950bb43.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### GRPO

Accuracy: 48.2. Result ID: 69a87e046dbb7075b950bb42.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### SFT

Accuracy: 43.9. Result ID: 69a87e046dbb7075b950bb41.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### APPO

Accuracy: 35. Result ID: 69a87e046dbb7075b950bb3f.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### APPO

Accuracy: 34.7. Result ID: 69a87e046dbb7075b950bb88.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"34K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

### DAPO

Accuracy: 31.8. Result ID: 69a87e046dbb7075b950bb3e.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### GRPO

Accuracy: 31. Result ID: 69a87e046dbb7075b950bb3d.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### Base Model

Accuracy: 29.6. Result ID: 69a87e046dbb7075b950bb40.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### Video-R1

Accuracy: 28.8. Result ID: 69a87e046dbb7075b950bb86.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c6f1e209131cea8e0ec.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"260K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

### SFT

Accuracy: 28.6. Result ID: 69a87e046dbb7075b950bb3c.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### VideoChat-R1

Accuracy: 27.8. Result ID: 69a87e046dbb7075b950bb84.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba11e209131cea89767.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"18K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

### VideoRFT

Accuracy: 27.7. Result ID: 69a87e046dbb7075b950bb87.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c241e209131cea8ca8e.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"310K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

### Base Model

Accuracy: 27. Result ID: 69a87e046dbb7075b950bb3b.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Resolution":"224x224","Sampling Rate":"1 fps","Max Frames":"60"}. This record does not assert independent reproduction.

### GRPO-CARE

Accuracy: 26. Result ID: 69a87e046dbb7075b950bb85.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79901e209131cea7e41e.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"260K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

### TW-GRPO

Accuracy: 25.6. Result ID: 69a87e046dbb7075b950bb83.

[Reporting paper](<https://www.sota2.com/research/paper/69a5195f2d603e6de3e3c777.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa793e1e209131cea7cd67.md>)

Recorded evaluation configuration: {"Size":"7B","Training Data":"1K","Resolution":"224 x 224","Sampling Rate (fps)":"1fps","Maximum Frames":"30","Zero-shot evaluation":"true"}. This record does not assert independent reproduction.

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