Summary
CARA (Concept-Aware Risk Attention) is an intrinsically interpretable spatio-temporal framework for collision anticipation in autonomous driving, presented in arXiv paper 2607.22494 by Zhishan Tao et al. The method addresses limitations of existing approaches: feature-driven models are opaque, post-hoc explanations lack fidelity, and prior concept-based methods target static recognition rather than dynamic driving scenes. CARA derives domain-grounded risk concepts from accident narratives, aligns them with video frames using vision-language similarity, and organizes them into evolving concept trajectories. These trajectories serve as explicit risk evidence that guides spatial attention, temporal attention, and anticipation, allowing semantic concepts to directly influence where the model attends and how risk is predicted over time. By treating semantic risk factors as dynamic intermediate evidence rather than auxiliary post-hoc explanations, CARA tightly couples interpretability with the prediction process. Experiments on three benchmarks show CARA consistently improves anticipation accuracy and warning lead time over strong baselines while providing sparse, semantically grounded concept evidence.
Overview
Field: Computer Vision
Authors: Zhishan Tao, Ruoyu Wang, Yucheng Wu, Enjun Du, Yilei Yuan, Sherwin Ho, Yue Su, Jinbo Su, Yi Hong
Published: 2026-07-24
arXiv: 2607.22494
Summary
Collision anticipation in autonomous driving requires not only accurate early warnings but also interpretable reasoning about what risk factors are being tracked and how risk evolves over time. Existing methods fall short in this regard: feature-driven models are opaque, post-hoc explanations often lack fidelity, and concept-based methods are mostly designed for static recognition rather than dynamic driving scenes.
The authors propose CARA (Concept-Aware Risk Attention), an intrinsically interpretable spatio-temporal framework for collision anticipation. CARA works as follows:
- Concept derivation: extracts domain-grounded risk concepts from accident narratives.
- Alignment: matches these concepts with video frames via vision-language similarity.
- Concept trajectories: organizes concepts into evolving trajectories over time.
These trajectories provide explicit risk evidence that guides spatial attention, temporal attention, and anticipation, allowing semantic concepts to directly influence where the model attends and how it predicts risk over time. By treating semantic risk factors as dynamic intermediate evidence rather than auxiliary post-hoc explanations, CARA tightly couples interpretability with the prediction process.
Results
Extensive experiments on three benchmarks demonstrate that CARA consistently improves anticipation accuracy and warning lead time, outperforming strong baselines while providing sparse and semantically grounded concept evidence.
---
*Auto-collected on 2026-07-28.*
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/178503745