arXiv Machine Learning
It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Aug 12
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
arXiv:2608. 10513v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones.
Hugging Face Trending Papers
Aug 11
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones. We propose SafeCap, a reinforcement-learning framework that aligns LVLMs through learned self-captioning.
arXiv AI
Jun 2
PaSBench-Video: A Streaming Video Benchmark for Proactive Safety Warning
arXiv:2606. 02443v1 Announce Type: cross Abstract: Between the first visible sign of danger and the moment an accident occurs, there is often a window where intervention remains possible.
arXiv AI
Jun 11
Task-Aligned Stability Analysis of Vision-Language Models for Autonomous Driving Hazard Detection
arXiv:2606. 11889v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly used for scene understanding in autonomous driving, but robustness analysis often relies on task-agnostic embedding stability alone.
arXiv AI
Jul 10
AUTOPILOT VQA: Benchmarking Vision-Language Models for Incident-Centric Dashcam Understanding
arXiv:2607. 08745v1 Announce Type: new Abstract: Recent advances in Vision-Language Models, Large Language Models, and Multimodal Large Language Models have improved autonomous driving tasks such as scene understanding, decision making, trajectory prediction, and visual question answering.
arXiv Machine Learning
2d ago
BD-HazardVLM: Probing Vision-Language Models for Latent Defensive-Driving Hazards in Bangladesh Road Scenes
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."