CrashDiffuser: VLM-Guided Collision Intent Reasoning for Fine-Grained Safety-Critical Traffic Scenario Generation

• URL: http://arxiv.org/abs/2609.02270v1
• Date: Wed, 02 Sep 2026 08:18:20 GMT
• Title: CrashDiffuser: VLM-Guided Collision Intent Reasoning for Fine-Grained Safety-Critical Traffic Scenario Generation
• Abstract summary: We study fine-grained safety-critical scenario generation, where success requires both a target collision and a specified head, rear, or side contact region.<n>We propose CrashDiffuser, a closed-loop VLM-guided diffusion framework that decouples semantic collision reasoning from continuous trajectory synthesis.<n>On WOMD-derived closed-loop scenarios, CrashDiffuser achieves a target-collision rate of 50.33% in a single attempt and 67.98% after three attempts.
• Score: 15.782911514743306
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generating safety-critical scenarios is essential for evaluating autonomous driving systems. However, existing generators primarily focus on inducing collisions and offer limited control over where contact occurs on the target vehicle. In this paper, we study fine-grained safety-critical scenario generation, where success requires both a target collision and a specified head, rear, or side contact region. We propose CrashDiffuser, a closed-loop VLM-guided diffusion framework that decouples semantic collision reasoning from continuous trajectory synthesis through a hierarchical collision-intent interface derived from the requested target contact region. At initialization, the VLM extracts reusable scene-level context; at each replanning step, it predicts a structured action tuple describing speed change, turning behavior, and collision stage. This intent conditions a diffusion model to generate executable adversarial trajectories, while collision-guided sampling, candidate selection, and short-horizon replanning adapt generation to the target vehicle's evolving behavior. On WOMD-derived closed-loop scenarios, CrashDiffuser achieves a target-collision rate of 50.33% in a single attempt and 67.98% after three attempts, together with a contact-region control success rate of 40.05% and competitive trajectory naturalness. Component ablations further support the proposed design.

Related papers

• Collision Snapshot Guided Time-Reversed Safety-Critical Scenario Generation [2.7044962986457945]
  We introduce underlineCOllision underlineSnapshot guided underlineE-underlineReversed safety-critical scenario generation (COSTER)<n>COSTER is a framework that leverages learned traffic priors to determine plausible collision times and locations.<n>Experiments show that COSTER outperforms existing methods in plausibility, diversity, and data efficiency.
  arXiv  Detail & Related papers  (2026-09-06T07:18:07Z)
• End-to-end Conditional Diffusion for Realistic and Controllable Visual Traffic Scenario Generation [59.188006170485046]
  E2E-CDiff is an end-to-end conditional diffusion framework for controllable and realistic scenario generation.<n>Differentiable guidance regulates speed, enforces drivable-area compliance, and supports collision-avoidance or collision-seeking behaviors.<n>E2E-CDiff also performs competitively as a learning-based ego planner.
  arXiv  Detail & Related papers  (2026-07-21T02:13:25Z)
• CCFM: Collision-Constrained Flow Matching for Safety-Critical Scenario Generation [12.01753684349527]
  We introduce Collision-Constrained Flow Matching (CCFM), a novel framework that guarantees precise collision control through hard physical constraints.<n>CCFM collision rate up to 46.4% on nuScenes and 83.1% on nuPlan, significantly outperforming baselines while preserving realistic driving behavior.
  arXiv  Detail & Related papers  (2026-07-05T18:26:14Z)
• From Attacks to Curricula: Learnability-Guided Adversarial Training for Safe Autonomous Driving [56.30087557121323]
  AlignADV is a learnability-guided closed-loop adversarial training framework.<n>It converts adversarial scenarios into resolvable and capability-aligned curricula.<n> Experiments show it reduces training steps by up to 40.6 percent.
  arXiv  Detail & Related papers  (2026-06-12T02:13:55Z)
• KG-ASG: Collision-Knowledge-Guided Closed-Loop Adversarial Scenario Generation With Primary-Support Attribution [11.204233852756808]
  KG-ASG is a collision-knowledge-guided closed-loop adversarial scenario generation framework with primary-support attribution.<n> Experiments on WOMD scenarios reconstructed in MetaDrive show that KG-ASG achieves strong adversarial effectiveness.
  arXiv  Detail & Related papers  (2026-05-17T09:04:48Z)
• Map-Agnostic And Interactive Safety-Critical Scenario Generation via Multi-Objective Tree Search [7.691544372846608]
  This paper presents a novel framework for traffic-flow level safety-critical scenario generation via Monte Carlo Tree Search.<n>A hybrid Upper Confidence Bound (UCB) and Lower Confidence Bound (LCB) search strategy is introduced to balance exploratory efficiency with risk-averse decision-making.<n>We validate our approach across four high-risk accident zones in Hong Kong's complex urban environments.
  arXiv  Detail & Related papers  (2026-03-04T12:19:02Z)
• Optimization-Guided Diffusion for Interactive Scene Generation [52.23368750264419]
  We present OMEGA, an optimization-guided, training-free framework that enforces structural consistency and interaction awareness during diffusion-based sampling.<n>We show that OMEGA improves generation realism, consistency, and controllability, increasing the ratio of physically and behaviorally valid scenes.<n>Our approach can also generate $5times$ more near-collision frames with a time-to-collision under three seconds.
  arXiv  Detail & Related papers  (2025-12-08T15:56:18Z)
• Controllable Collision Scenario Generation via Collision Pattern Prediction [22.736678908526653]
  We introduce controllable collision scenario generation, where the goal is to produce trajectories that realize a user-specified collision type and time-to-accident (TTA)<n>We propose a framework that predicts Collision Pattern, a compact and interpretable representation that captures the spatial configuration of the ego and the adversarial vehicles at impact.<n>Experiments show that our approach outperforms strong baselines in both collision rate and controllability.
  arXiv  Detail & Related papers  (2025-10-14T06:56:33Z)
• Building a Foundational Guardrail for General Agentic Systems via Synthetic Data [76.18834864749606]
  LLM agents can plan multi-step tasks, intervening at the planning stage-before any action is executed-is often the safest way to prevent harm.<n>Existing guardrails mostly operate post-execution, which is difficult to scale and leaves little room for controllable supervision at the plan level.<n>We introduce AuraGen, a controllable engine that synthesizes benign trajectories, injects category-labeled risks with difficulty, and filters outputs via an automated reward model.
  arXiv  Detail & Related papers  (2025-10-10T18:42:32Z)
• ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving [49.07731497951963]
  ReCogDrive is a novel Reinforced Cognitive framework for end-to-end autonomous driving.<n>We introduce a hierarchical data pipeline that mimics the sequential cognitive process of human drivers.<n>We then address the language-action mismatch by injecting the VLM's learned driving priors into a diffusion planner.
  arXiv  Detail & Related papers  (2025-06-09T03:14:04Z)
• Generating Useful Accident-Prone Driving Scenarios via a Learned Traffic Prior [135.78858513845233]
  STRIVE is a method to automatically generate challenging scenarios that cause a given planner to produce undesirable behavior, like collisions. To maintain scenario plausibility, the key idea is to leverage a learned model of traffic motion in the form of a graph-based conditional VAE. A subsequent optimization is used to find a "solution" to the scenario, ensuring it is useful to improve the given planner.
  arXiv  Detail & Related papers  (2021-12-09T18:03:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.