BD-HazardVLM: Probing Vision-Language Models for Latent Defensive-Driving Hazards in Bangladesh Road Scenes
Read the original on arXiv Machine Learning →BD-HazardVLM is a diagnostic benchmark of 500 Bangladesh road-scene images designed to probe vision‑language models’ defensive‑driving reasoning in weakly structured Global South traffic. The benchmark includes annotations for hazard presence, category, risk timing, severity, defensive actions, visible evidence, and bias traps, and evaluates models such as Qwen2.5‑VL‑7B, InternVL3‑8B, and MiniCPM‑V‑4.6. Results show high binary hazard‑presence accuracy but poor fine‑grained reasoning, with category accuracy below 40 %, risk‑timing accuracy under 13 %, and defensive‑action matching under 2 %. whyItMatters":"The study demonstrates that coarse hazard recognition does not guarantee reliable fine‑grained defensive‑driving reasoning, highlighting the need for geographically diverse safety evaluation of vision‑language models."
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.