PaveVQA - Quantification MAPE: leaderboard
PaveVQA is the vision-language question answering part of PaveBench (2026), built on real highway pavement inspection images with single-turn, multi-turn and expert-corrected questions about distress recognition, localization, quantification and maintenance. This board is the mean absolute percentage error of numerical answers such as distress area, length and severity, answered zero-shot, with no upper limit; lower is better.
Metric: Mean absolute percentage error (%, no upper limit) of quantitative answers (distress area and length measures), of the zero-shot model on PaveVQA, the vision-language question answering part of PaveBench (real highway pavement images, single-turn, multi-turn and expert-corrected questions); lower is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 3 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 VL 3B | 116.2 | #2639 |
| 2 | LLaVA-OneVision-7B | 158.24 | #2981 |
| 3 | DeepSeek-VL2-Small | 531.21 | #2464 |
Interactive version: theaggregate.ai/benchmark?slug=pavevqa-quantification-mape · How It Works · Data refreshed daily, snapshot 2026-11-21.