# Video Question Answering on LongVideoBench (LVB)

Source: [Video Question Answering on LongVideoBench (LVB)](<https://www.sota2.com/research/sota/video-question-answering-on-longvideobench-lvb>)

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

Data updated: 2026-07-03T07:10:42.813Z

Evaluation of model performance on the LongVideoBench (LVB) dataset for Video Question Answering tasks, measured using 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 Question Answering |
| Dataset (including any stated split) | LongVideoBench (LVB) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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-longvideobench-lvb.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2-VL + ReQuest | 66.3 | Not recorded | 6a47420b143ebdcacacc5c6e |
| Qwen2-VL | 62.7 | Not recorded | 6a47420b143ebdcacacc5c6e |
| LLaVa-OneVision + ReQuest | 60.2 | Not recorded | 6a47420b143ebdcacacc5c6e |
| LLaVA-Video + ReQuest | 60.1 | Not recorded | 6a47420b143ebdcacacc5c6e |
| LLaVA-Video | 58 | Not recorded | 6a47420b143ebdcacacc5c6e |
| NVILA | 57.7 | Not recorded | 6a47420b143ebdcacacc5c6e |
| LLaVa-OneVision | 56.6 | Not recorded | 6a47420b143ebdcacacc5c6e |
| Video-LLaVA | 39.1 | Not recorded | 6a47420b143ebdcacacc5c6e |

### Qwen2-VL + ReQuest

Accuracy: 66.3. Result ID: 6a47607160b88bb339928619.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

Recorded evaluation configuration: {"LLM Size":"8B","#Frames":"≤ 512","zero-shot transfer":"true"}. This record does not assert independent reproduction.

### Qwen2-VL

Accuracy: 62.7. Result ID: 6a47607160b88bb339928618.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"LLM Size":"8B","#Frames":"512","reproduced":"true"}. This record does not assert independent reproduction.

### LLaVa-OneVision + ReQuest

Accuracy: 60.2. Result ID: 6a47607160b88bb339928617.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"32","zero-shot transfer":"true"}. This record does not assert independent reproduction.

### LLaVA-Video + ReQuest

Accuracy: 60.1. Result ID: 6a47607160b88bb339928615.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"32"}. This record does not assert independent reproduction.

### LLaVA-Video

Accuracy: 58. Result ID: 6a47607160b88bb339928614.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

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

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

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"32","reproduced":"true"}. This record does not assert independent reproduction.

### NVILA

Accuracy: 57.7. Result ID: 6a47607160b88bb339928613.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77481e209131cea723b2.md>)

Recorded evaluation configuration: {"LLM Size":"8B","#Frames":"1024"}. This record does not assert independent reproduction.

### LLaVa-OneVision

Accuracy: 56.6. Result ID: 6a47607160b88bb339928616.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

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

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

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"32","reproduced":"true"}. This record does not assert independent reproduction.

### Video-LLaVA

Accuracy: 39.1. Result ID: 6a47607160b88bb339928612.

[Reporting paper](<https://www.sota2.com/research/paper/6a47420b143ebdcacacc5c6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65534.md>)

[Implementation](<https://github.com/PKU-YuanGroup/Video-LLaVA>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"8"}. This record does not assert independent reproduction.

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