# CrashSight: A Phase-Aware, Infrastructure-Centric Video Benchmark for Traffic Crash Scene Understanding and Reasoning

arXiv: https://arxiv.org/abs/2604.08457 · HTML: https://mlsanity.com/paper/2604.08457/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 15 of its 37 references that are in the map.
- First posted on arXiv 9 Apr 2026.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)

## Builds on (most cited first, 15 of 15)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [DeepSeek-OCR: Contexts Optical Compression](https://mlsanity.com/paper/2510.18234/) (2025)
- [PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model](https://mlsanity.com/paper/2510.14528/) (2025)
- [Improving LLM Video Understanding with 16 Frames Per Second](https://mlsanity.com/paper/2503.13956/) (2025)
- [TUMTraffic-VideoQA: A Benchmark for Unified Spatio-Temporal Video Understanding in Traffic Scenes](https://mlsanity.com/paper/2502.02449/) (2025)
- [SurveillanceVQA-589K: A Benchmark for Comprehensive Surveillance Video-Language Understanding with Large Models](https://mlsanity.com/paper/2505.12589/) (2025)
- [Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence](https://mlsanity.com/paper/2510.20470/) (2025)
- [DriveQA: Passing the Driving Knowledge Test](https://mlsanity.com/paper/2508.21824/) (2025)
- [SafePLUG: Empowering Multimodal LLMs with Pixel-Level Insight and Temporal Grounding for Traffic Accident Understanding](https://mlsanity.com/paper/2508.06763/) (2025)
- [CoT-VLM4Tar: Chain-of-Thought Guided Vision-Language Models for Traffic Anomaly Resolution](https://mlsanity.com/paper/2503.01632/) (2025)
- [InterAct-Video: Reasoning-Rich Video QA for Urban Traffic](https://mlsanity.com/paper/2507.14743/) (2025)

## Built on by (most cited first, 4 of 4)

- [An Agentic Workflow for Detecting Personally Identifiable Information in Crash Narratives](https://mlsanity.com/paper/2604.15369/) (2026)
- [DispatchRAG: Grounding Emergency Dispatch Decisions in Real-World Protocols from Traffic Accident Video](https://mlsanity.com/paper/2607.23132/) (2026)
- [Forecasting the Emergence and Evolution of Crash Hotspots: A Unified Deep Learning Framework for Proactive Traffic Safety](https://mlsanity.com/paper/2607.24168/) (2026)
- [Zero-Shot Traffic Accident Detection via a Coarse-to-Fine VLM-Tracking Pipeline](https://mlsanity.com/paper/2608.08867/) (2026)

## Abstract (arXiv)

Cooperative autonomous driving requires traffic scene understanding from both vehicle and infrastructure perspectives. While vision-language models (VLMs) show strong general reasoning capabilities, their performance in safety-critical traffic scenarios remains insufficiently evaluated due to the ego-vehicle focus of existing benchmarks. To bridge this gap, we present CrashSight, a large-scale vision-language benchmark for roadway crash understanding using real-world roadside camera data. The dataset comprises 250 crash videos, annotated with 13K multiple-choice question-answer pairs organized under a two-tier taxonomy. Tier 1 evaluates the visual grounding of scene context and involved parties, while Tier 2 probes higher-level reasoning, including crash mechanics, causal attribution, temporal progression, and post-crash outcomes. We benchmark 8 state-of-the-art VLMs and show that, despite strong scene description capabilities, current models struggle with temporal and causal reasoning in safety-critical scenarios. We provide a detailed analysis of failure scenarios and discuss directions for improving VLM crash understanding. The benchmark provides a standardized evaluation framework for infrastructure-assisted perception in cooperative autonomous driving. The CrashSight benchmark, including the full dataset and code, is accessible at https://mcgrche.github.io/crashsight.