# Driving Video Question Answering on LingoQA 25% Token Retention

Source: [Driving Video Question Answering on LingoQA 25% Token Retention](<https://www.sota2.com/research/sota/driving-video-question-answering-on-lingoqa-25-percent-token-retention>)

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

Data updated: 2026-04-23T08:08:56.189Z

Evaluation of model performance on the LingoQA dataset subset constrained to 25% token retention, focusing on Driving Video Question Answering, measured using the LingoJudge metric.

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 | Driving Video Question Answering |
| Dataset (including any stated split) | LingoQA 25% Token Retention |
| Metric | LingoJudge Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [LingoJudge Score](<https://www.sota2.com/research/sota/driving-video-question-answering-on-lingoqa-25-percent-token-retention.md?metric=LingoJudge%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | LingoJudge Score | Result date | Report paper |
| --- | --- | --- | --- |
| DriveMM (Vanilla) | 70.8 | Not recorded | 69e85ba27d0172140ec80f94 |
| ST-Prune | 68.2 | Not recorded | 69e85ba27d0172140ec80f94 |
| VisPruner | 65.4 | Not recorded | 69e85ba27d0172140ec80f94 |
| PACT | 64 | Not recorded | 69e85ba27d0172140ec80f94 |
| DivPrune | 62.8 | Not recorded | 69e85ba27d0172140ec80f94 |

### DriveMM (Vanilla)

LingoJudge Score: 70.8. Result ID: 69e9d39747f34ce924d0d94f.

[Reporting paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a741e209131cea835a8.md>)

Recorded evaluation configuration: {"Token Retention Rate":"Full Tokens"}. This record does not assert independent reproduction.

### ST-Prune

LingoJudge Score: 68.2. Result ID: 69e9d39747f34ce924d0d953.

[Reporting paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

Recorded evaluation configuration: {"Token Retention Rate":"25%"}. This record does not assert independent reproduction.

### VisPruner

LingoJudge Score: 65.4. Result ID: 69e9d39747f34ce924d0d950.

[Reporting paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78741e209131cea786bf.md>)

Recorded evaluation configuration: {"Token Retention Rate":"25%","Venue":"ICCV 2025"}. This record does not assert independent reproduction.

### PACT

LingoJudge Score: 64. Result ID: 69e9d39747f34ce924d0d952.

[Reporting paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ec1e209131cea76055.md>)

Recorded evaluation configuration: {"Token Retention Rate":"25%","Venue":"CVPR 2025"}. This record does not assert independent reproduction.

### DivPrune

LingoJudge Score: 62.8. Result ID: 69e9d39747f34ce924d0d951.

[Reporting paper](<https://www.sota2.com/research/paper/69e85ba27d0172140ec80f94.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c41e209131cea7532a.md>)

Recorded evaluation configuration: {"Token Retention Rate":"25%","Venue":"CVPR 2025"}. This record does not assert independent reproduction.

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