# Video Question Answering on VideoMME with subtitles

Source: [Video Question Answering on VideoMME with subtitles](<https://www.sota2.com/research/sota/video-question-answering-on-videomme-with-subtitles>)

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

Data updated: 2026-02-26T05:26:51.749Z

Evaluation of model performance on the VideoMME dataset, specifically focusing on samples containing subtitles, for Video Question Answering, reporting scores categorized by question length (short, medium, long).

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) | VideoMME with subtitles |
| Metric | Acc (Overall) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Acc (Overall)](<https://www.sota2.com/research/sota/video-question-answering-on-videomme-with-subtitles.md?metric=Acc%20(Overall)>): Higher is better; unit: PERCENT.

- [Acc (Short)](<https://www.sota2.com/research/sota/video-question-answering-on-videomme-with-subtitles.md?metric=Acc%20(Short)>): Higher is better; unit: PERCENT.

- [Acc (Medium)](<https://www.sota2.com/research/sota/video-question-answering-on-videomme-with-subtitles.md?metric=Acc%20(Medium)>): Higher is better; unit: PERCENT.

- [Acc (Long)](<https://www.sota2.com/research/sota/video-question-answering-on-videomme-with-subtitles.md?metric=Acc%20(Long)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Acc (Overall) | Result date | Report paper |
| --- | --- | --- | --- |
| HICom (Ours) | 64.8 | Not recorded | 68fa77801e209131cea739ee |
| HICom (Ours) | 63.3 | Not recorded | 68fa77801e209131cea739ee |
| LLaVA-Video | 62.9 | Not recorded | 68fa77801e209131cea739ee |
| PAVE-7B (w/ video feature) | 62.9 | Not recorded | 68fa779f1e209131cea746cf |
| LLaVA-OneVision | 61.9 | Not recorded | 68fa77801e209131cea739ee |
| LLaVA-OV-7B | 61.5 | Not recorded | 68fa779f1e209131cea746cf |
| HICom (Ours) | 60.1 | Not recorded | 68fa77801e209131cea739ee |
| VITA | 59.2 | Not recorded | 68fa77801e209131cea739ee |
| HICom (Ours) | 55.7 | Not recorded | 68fa77801e209131cea739ee |
| LongVA | 54.3 | Not recorded | 68fa77801e209131cea739ee |
| PAVE-0.5B (w/ video feature) | 48.6 | Not recorded | 68fa779f1e209131cea746cf |
| Chat-Univi-1.5 | 45.9 | Not recorded | 68fa77801e209131cea739ee |
| VideoChat2-Mistral | 43.8 | Not recorded | 68fa77801e209131cea739ee |
| LLaVA-OV-0.5B | 43.5 | Not recorded | 68fa779f1e209131cea746cf |
| Video-LLaVA | 41.6 | Not recorded | 68fa77801e209131cea739ee |

### HICom (Ours)

Acc (Overall): 64.8. Result ID: 699498333a6c3148491e4297.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Implementation](<https://github.com/lntzm/HICom>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"128","#Tokens":"2624","inference\_frames\_sampling\_32":true}. This record does not assert independent reproduction.

### HICom (Ours)

Acc (Overall): 63.3. Result ID: 699498333a6c3148491e4296.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Implementation](<https://github.com/lntzm/HICom>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"64","#Tokens":"1328","inference\_frames\_sampling\_32":true}. This record does not assert independent reproduction.

### LLaVA-Video

Acc (Overall): 62.9. Result ID: 699498333a6c3148491e4293.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.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","#Tokens":"6272"}. This record does not assert independent reproduction.

### PAVE-7B (w/ video feature)

Acc (Overall): 62.9. Result ID: 6995554b3badfaffe645b21f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

[Implementation](<https://github.com/dragonlzm/PAVE>)

Recorded evaluation configuration: {"FLOPs (TB)":"98.63","Total/Trainable Params":"8.2B/170.5M"}. This record does not assert independent reproduction.

### LLaVA-OneVision

Acc (Overall): 61.9. Result ID: 699498333a6c3148491e4292.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.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","#Tokens":"6272"}. This record does not assert independent reproduction.

### LLaVA-OV-7B

Acc (Overall): 61.5. Result ID: 6995554b3badfaffe645b21d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

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

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

Recorded evaluation configuration: {"FLOPs (TB)":"98.53","Total/Trainable Params":"8.2B/-"}. This record does not assert independent reproduction.

### HICom (Ours)

Acc (Overall): 60.1. Result ID: 699498333a6c3148491e4295.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Implementation](<https://github.com/lntzm/HICom>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"32","#Tokens":"680"}. This record does not assert independent reproduction.

### VITA

Acc (Overall): 59.2. Result ID: 699498333a6c3148491e4291.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79041e209131cea7bb11.md>)

Recorded evaluation configuration: {"LLM Size":"8x7B","#Frames":"16","#Tokens":"4096"}. This record does not assert independent reproduction.

### HICom (Ours)

Acc (Overall): 55.7. Result ID: 699498333a6c3148491e4294.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Implementation](<https://github.com/lntzm/HICom>)

Recorded evaluation configuration: {"LLM Size":"1.5B","#Frames":"32","#Tokens":"680"}. This record does not assert independent reproduction.

### LongVA

Acc (Overall): 54.3. Result ID: 699498333a6c3148491e4290.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ea1e209131cea5edfe.md>)

[Implementation](<https://github.com/EvolvingLMMs-Lab/LongVA>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"128","#Tokens":"18432"}. This record does not assert independent reproduction.

### PAVE-0.5B (w/ video feature)

Acc (Overall): 48.6. Result ID: 6995554b3badfaffe645b21e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

[Implementation](<https://github.com/dragonlzm/PAVE>)

Recorded evaluation configuration: {"FLOPs (TB)":"8.08","Total/Trainable Params":"0.9B/41.4M"}. This record does not assert independent reproduction.

### Chat-Univi-1.5

Acc (Overall): 45.9. Result ID: 699498333a6c3148491e428f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d31e209131cea58bff.md>)

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

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"64","#Tokens":"448"}. This record does not assert independent reproduction.

### VideoChat2-Mistral

Acc (Overall): 43.8. Result ID: 699498333a6c3148491e428e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73411e209131cea5b4b7.md>)

Recorded evaluation configuration: {"LLM Size":"7B","#Frames":"16","#Tokens":"1536"}. This record does not assert independent reproduction.

### LLaVA-OV-0.5B

Acc (Overall): 43.5. Result ID: 6995554b3badfaffe645b21c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cf.md>)

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

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

Recorded evaluation configuration: {"FLOPs (TB)":"8.01","Total/Trainable Params":"0.9B/-"}. This record does not assert independent reproduction.

### Video-LLaVA

Acc (Overall): 41.6. Result ID: 699498333a6c3148491e428d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77801e209131cea739ee.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","#Tokens":"2048"}. This record does not assert independent reproduction.

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