Research papers · Posted 2026.05.16 · Preprint

DriveSafe: A Framework for Risk Detection and Safety Suggestions in Driving Scenarios

Sainithin Artham, Shankar Gangisetty, Avijit Dasgupta, C. V. Jawahar · arXiv · English

Overview

DriveSafe grounds risk assessment in structured scene captions containing motion, depth and spatial cues. A lightweight adapter adds driving-specific knowledge, supporting hazard localization and safety suggestions evaluated on the DRAMA benchmark.

Author-supplied source information

Author-supplied information on arXiv.

8 pages

Version: v1

DriveSafe: A Framework for Risk Detection and Safety Suggestions in Driving Scenarios

Overview

Author-supplied source information

Participation enquiry

Your enquiry has been sent.

DriveSafe: A Framework for Risk Detection and Safety Suggestions in Driving Scenarios

Overview

Author-supplied source information

Related resources

Assuring Learning-Enabled Cyber-Physical Systems with Foundation Models: A Neuro-Symbolic Agenda for Testing, Verification, and Runtime Assurance

Expand-and-Prune: Test-Time Candidate-Space Adaptation for End-to-End Autonomous Driving

ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Prediction