RiskFlow: Fast and Faithful Safety-Critical Traffic Scenario Generation

• URL: http://arxiv.org/abs/2606.06423v1
• Date: Thu, 04 Jun 2026 17:28:42 GMT
• Title: RiskFlow: Fast and Faithful Safety-Critical Traffic Scenario Generation
• Abstract summary: RiskFlow is a closed-loop safety-critical multi-agent traffic generation framework.<n>It formulates future trajectory generation as transport in the action space.<n>It consistently improves realism while maintaining competitive safety-critical generation capability.
• Score: 17.00557757061364
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Safety-critical traffic scenario generation is essential for evaluating autonomous driving systems under rare but high-risk interactions. Existing diffusion-based methods offer strong controllability in closed-loop generation, but their iterative denoising process is computationally expensive and may accumulate sampling and guidance errors over long rollouts, causing unrealistic motion artifacts such as jitter, abnormal acceleration, and off-road behavior. To address these issues, we propose RiskFlow, a closed-loop safety-critical multi-agent traffic generation framework that formulates future trajectory generation as transport in the action space. Instead of relying on iterative denoising, RiskFlow learns an average velocity field over a finite interval to transform Gaussian action sequences into future acceleration and yaw-rate commands with a single forward pass, using a JVP-based objective for efficient and stable training. At test time, RiskFlow applies output-space guidance to the generated actions, steering selected critical agents toward risky interactions while regularizing off-road behavior, and reconstructs physically feasible trajectories through vehicle dynamics. Experiments on nuScenes with tbsim closed-loop evaluation show that RiskFlow achieves a strong adversariality-realism trade-off across multi-agent and long-horizon settings. Compared with representative baselines, RiskFlow consistently improves realism while maintaining competitive safety-critical generation capability, and substantially reduces inference time for evaluation.

Related papers

• RiskTraf: Risk-Extrapolated Residual Learning for Multi-Variate Traffic Flow Prediction [8.558982321151184]
  We develop a public benchmark suite that preserves raw flow, speed, and occupancy measurements from PeMS detectors for flow prediction.<n>We also propose textbfRiskTraf, a model-agnostic risk-extrapolated residual plug-in.<n>RiskTraf consistently improves diverse forecasting backbones and outperforms debiasing and distribution-shift adaptation methods.
  arXiv  Detail & Related papers  (2026-08-21T01:26:15Z)
• Anticipatory Risk-Guided Reinforcement Learning for Safe Flight Through Dynamic Clutter [21.995071172604153]
  We propose an anticipatory risk- reinforcement learning framework for quadrotors.<n>We use privileged simulator states to construct a directionally aligned future collision risk map.<n>Through an asymmetric actor-critic architecture, the network is trained to self-predict this structured risk, which explicitly guides the visual policy during deployment.<n>Our method effectively improves safety margins and flight efficiency in dense dynamic clutters compared to existing baselines.
  arXiv  Detail & Related papers  (2026-07-26T09:31:49Z)
• 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)
• Goal-Oriented Reactive Simulation for Closed-Loop Trajectory Prediction [9.812666469580872]
  Current trajectory prediction models are primarily trained in an open-loop manner.<n>We propose an on-policy closed-loop training paradigm optimized for high-frequency, receding horizon ego prediction.
  arXiv  Detail & Related papers  (2026-03-25T10:24:16Z)
• SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gating [5.252722148826105]
  SafeFlow is a text-driven humanoid whole-body control framework.<n>It combines physics-guided motion generation with a 3-Stage Safety Gate driven by explicit risk indicators.<n>It outperforms prior diffusion-based methods in success rate, physical compliance, and inference speed.
  arXiv  Detail & Related papers  (2026-03-25T06:24:16Z)
• FASTER: Rethinking Real-Time Flow VLAs [82.58822112377923]
  Real-time execution is crucial for deploying Vision-Language-Action (VLA) models in the physical world.<n>We show that reaction time follows a uniform distribution determined jointly by the Time to First Action (TTFA) and the execution horizon.<n>We propose Fast Action Sampling for ImmediaTE Reaction (FASTER) to overcome this issue.
  arXiv  Detail & Related papers  (2026-03-19T17:51:37Z)
• Conditional Flow Matching for Continuous Anomaly Detection in Autonomous Driving on a Manifold-Aware Spectral Space [0.0]
  We present Deep-Flow, an unsupervised framework for safety-critical anomaly detection.<n>Unlike standard generative approaches that operate in unstable, high-dimensional coordinate spaces, Deep-Flow constrains the generative process to a low-rank spectral manifold.<n>This ensures kinematic smoothness by design and enables the computation of numerically stable, deterministic log-likelihood estimation.
  arXiv  Detail & Related papers  (2026-02-19T18:10:16Z)
• ARTIS: Agentic Risk-Aware Test-Time Scaling via Iterative Simulation [72.78362530982109]
  ARTIS, Agentic Risk-Aware Test-Time Scaling via Iterative Simulation, is a framework that decouples exploration from commitment.<n>We show that naive LLM-based simulators struggle to capture rare but high-impact failure modes.<n>We introduce a risk-aware tool simulator that emphasizes fidelity on failure-inducing actions.
  arXiv  Detail & Related papers  (2026-02-02T06:33:22Z)
• RIFT: Group-Relative RL Fine-Tuning for Realistic and Controllable Traffic Simulation [13.319344167881383]
  We introduce a dual-stage AV-centric simulation framework that conducts imitation learning pre-training in a data-driven simulator.<n>We then learn fine-tuning in a physics-based simulator to enhance style-level controllability.<n>In the fine-tuning stage, we propose RIFT, a novel group-relative RL fine-tuning strategy.
  arXiv  Detail & Related papers  (2025-05-06T09:12:37Z)
• Hydra-NeXt: Robust Closed-Loop Driving with Open-Loop Training [64.16445087751039]
  Hydra-NeXt is a novel multi-branch planning framework that unifies trajectory prediction, control prediction, and a trajectory refinement network in one model.<n> Hydra-NeXt surpasses the previous state-of-the-art by 22.98 DS and 17.49 SR, marking a significant advancement in autonomous driving.
  arXiv  Detail & Related papers  (2025-03-15T07:42:27Z)
• DeepMF: Deep Motion Factorization for Closed-Loop Safety-Critical Driving Scenario Simulation [11.059102404333885]
  Safety-critical traffic scenarios are of great practical relevance to evaluating the robustness of autonomous driving systems.<n>Existing algorithms for generating safety-critical scenarios rely on snippets of previously recorded traffic events.<n>In this paper, we propose the Deep Motion Factorization framework, which extends static safety-critical driving scenario generation to closed-loop and interactive adversarial traffic simulation.
  arXiv  Detail & Related papers  (2024-12-23T11:30:24Z)
• SAFE-SIM: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries [94.84458417662407]
  We introduce SAFE-SIM, a controllable closed-loop safety-critical simulation framework. Our approach yields two distinct advantages: 1) generating realistic long-tail safety-critical scenarios that closely reflect real-world conditions, and 2) providing controllable adversarial behavior for more comprehensive and interactive evaluations. We validate our framework empirically using the nuScenes and nuPlan datasets across multiple planners, demonstrating improvements in both realism and controllability.
  arXiv  Detail & Related papers  (2023-12-31T04:14:43Z)
• CAT: Closed-loop Adversarial Training for Safe End-to-End Driving [54.60865656161679]
  Adversarial Training (CAT) is a framework for safe end-to-end driving in autonomous vehicles. Cat aims to continuously improve the safety of driving agents by training the agent on safety-critical scenarios. Cat can effectively generate adversarial scenarios countering the agent being trained.
  arXiv  Detail & Related papers  (2023-10-19T02:49:31Z)

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.