# Video Question Answering on VideoEspresso (test)

Source: [Video Question Answering on VideoEspresso (test)](<https://www.sota2.com/research/sota/video-question-answering-on-videoespresso-test>)

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

Data updated: 2026-02-26T05:32:53.141Z

Evaluation of model performance on the test split of the VideoEspresso dataset for Video Question Answering. Metrics include Accuracy, Frames, Rounds, and Time.

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 Question Answering |
| Dataset (including any stated split) | VideoEspresso (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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-question-answering-on-videoespresso-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Frames](<https://www.sota2.com/research/sota/video-question-answering-on-videoespresso-test.md?metric=Frames>): Higher is better; unit: SCALAR.

- [Rounds](<https://www.sota2.com/research/sota/video-question-answering-on-videoespresso-test.md?metric=Rounds>): Higher is better; unit: SCALAR.

- [Time (s)](<https://www.sota2.com/research/sota/video-question-answering-on-videoespresso-test.md?metric=Time%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| REVISE (Reinforced Fine-Tuning) | 0.278 | Not recorded | 6993f5e79ebc499c974098f3 |
| Supervised Format Fine-Tuning | 0.213 | Not recorded | 6993f5e79ebc499c974098f3 |
| Plug-and-Play | 0.201 | Not recorded | 6993f5e79ebc499c974098f3 |
| Direct Reasoning | 0.126 | Not recorded | 6993f5e79ebc499c974098f3 |

### REVISE (Reinforced Fine-Tuning)

Accuracy: 0.278. Result ID: 699a7e315698665ec407b32b.

[Reporting paper](<https://www.sota2.com/research/paper/6993f5e79ebc499c974098f3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6993f5e79ebc499c974098f3.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-VL-3B","Reasoning Category":"Reinforced Fine-Tuning (RFT)"}. This record does not assert independent reproduction.

### Supervised Format Fine-Tuning

Accuracy: 0.213. Result ID: 699a7e315698665ec407b32a.

[Reporting paper](<https://www.sota2.com/research/paper/6993f5e79ebc499c974098f3.md>)

Recorded evaluation configuration: {"Reasoning Category":"supervised format fine-tuning"}. This record does not assert independent reproduction.

### Plug-and-Play

Accuracy: 0.201. Result ID: 699a7e315698665ec407b329.

[Reporting paper](<https://www.sota2.com/research/paper/6993f5e79ebc499c974098f3.md>)

Recorded evaluation configuration: {"Reasoning Category":"plug-and-play without updates"}. This record does not assert independent reproduction.

### Direct Reasoning

Accuracy: 0.126. Result ID: 699a7e315698665ec407b328.

[Reporting paper](<https://www.sota2.com/research/paper/6993f5e79ebc499c974098f3.md>)

Recorded evaluation configuration: {"Reasoning Category":"direct video-understanding"}. This record does not assert independent reproduction.

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