CrashSight - Robustness: leaderboard
Hallucination probe questions (cannot-determine, absent-entity, temporal-trap and false-premise items) of CrashSight (2026), a multiple-choice VQA benchmark on real roadside-camera traffic crash videos answered zero-shot from four frames. Accuracy in percent; higher is better.
Metric: Accuracy (%) on hallucination-probe questions (cannot-determine, absent-entity, temporal-trap and false-premise items whose answer resists fabrication), four-option multiple-choice VQA on held-out roadside-camera crash videos of CrashSight-VQA (4 frames at 1 fps, zero-shot, deterministic decoding, exact letter match); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3-8B | 82.2 | #1782 |
| 2 | Qwen 2.5 VL 7B Instruct | 73.3 | #1902 |
| 3 | Qwen 2.5 VL 3B Instruct | 71.1 | #2713 |
| 4 | InternVL3-2B | 71.1 | #3290 |
| 5 | LLaVA-NeXT-Video-7B | 66.1 | #5734 |
| 6 | LLaVA-OneVision-0.5B | 18.9 | #7956 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=crashsight-robustness · How It Works · Data refreshed daily, snapshot 2026-11-21.