mlsanityData to 14 Oct 2026

Flow-ERD: Agent-type Aware Flow Matching with Entropy-Regularized Distillation for Diverse Traffic Simulation

arXiv 2607.06957 ↗First posted 8 Jul 2026cs.RO cs.LGcited by 0

On the map: Diffusion models → Diffusion and flow policies for robotics

Abstract (arXiv)

Realistic and diverse traffic simulation is essential to autonomous driving development. Yet prevailing benchmarks predominantly reward realism, and recent methods have optimized accordingly, leaving diversity underexplored. We introduce Flow-ERD, a multi-agent simulator that pursues realism and diversity jointly. Its backbone, Agent-Type Aware Flow Matching (AFM), couples flow matching's multi-modal expressiveness with type-specific kinematic execution. It preserves fine-grained diversity while keeping motions consistent with each agent type. A second stage, Entropy-Regularized Distillation (ERD), fine-tunes the closed-loop rollout distribution with an entropy-regularized reverse-KL objective. This mitigates covariate shift while explicitly preventing collapse onto high-density modes. We evaluate Flow-ERD with a log-free diversity metric alongside standard realism scores. Flow-ERD ranks first on the WOSAC test benchmark and dominates the realism--diversity Pareto front among reproducible baselines. Our project page is available hrefhttps://seulbinhwang.github.io/flow-erd-project-page/here.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 31 of its 40 references that are in the map. First posted on arXiv 8 Jul 2026.

Builds on

31 of 40 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025LANGTRAJ: Diffusion Model and Dataset for Language-Conditioned Trajectory Simulation15 Apr
  2. 2024SceneDiffuser: Efficient and Controllable Driving Simulation Initialization and Rollout5 Dec
  3. 2024DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving22 Nov
  4. 2023SAFE-SIM: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries31 Dec
  5. 2022Guided Conditional Diffusion for Controllable Traffic Simulation31 Oct

Other references · 26

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2022Scalable Diffusion Models with Transformerscited by 6,054
  3. 2022Classifier-Free Diffusion Guidancecited by 5,847
  4. 2022Flow Matching for Generative Modelingcited by 5,702
  5. 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
  6. 2016Generative Adversarial Imitation Learningcited by 1,597
  7. 2023One-step Diffusion with Distribution Matching Distillationcited by 1,019
  8. 2025Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusioncited by 625
  9. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 537
  10. 2016A Survey of Motion Planning and Control Techniques for Self-driving Urban Vehiclescited by 254
  11. 2021TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviorscited by 193
  12. 2023Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Researchcited by 160
14 more
  1. 2019Imitation Learning as f-Divergence Minimizationcited by 123
  2. 2023The Waymo Open Sim Agents Challengecited by 86
  3. 2023Trajeglish: Traffic Modeling as Next-Token Predictioncited by 60
  4. 2024BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Predictioncited by 50
  5. 2024Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Modelscited by 40
  6. 2024Versatile Behavior Diffusion for Generalized Traffic Agent Simulationcited by 30
  7. 2024KiGRAS: Kinematic-Driven Generative Model for Realistic Agent Simulationcited by 17
  8. 2025Advancing Multi-agent Traffic Simulation via R1-Style Reinforcement Fine-Tuningcited by 14
  9. 2025Decoupled Diffusion Sparks Adaptive Scene Generationcited by 12
  10. 2018Guaranteeing Consistency in a Motion Planning and Control Architecture Using a Kinematic Bicycle Modelcited by 6
  11. 2025RoaD: Rollouts as Demonstrations for Closed-Loop Supervised Fine-Tuning of Autonomous Driving Policiescited by 6
  12. 2025DecompGAIL: Learning Realistic Traffic Behaviors with Decomposed Multi-Agent Generative Adversarial Imitation Learningcited by 5
  13. 2025MDG: Masked Denoising Generation for Multi-Agent Behavior Modeling in Traffic Environmentscited by 4
  14. 2026RLFTSim: Realistic and Controllable Multi-Agent Traffic Simulation via Reinforcement Learning Fine-Tuningcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024TrafficGamer: Reliable and Flexible Traffic Simulation for Safety-Critical Scenarios with Game-Theoretic Oracles
  2. 2026Learning Direct Control Policies with Flow Matching for Autonomous Driving
  3. 2026ForSim: Stepwise Forward Simulation for Traffic Policy Fine-Tuning
  4. 2026Real-Time Generative Policy via Langevin-Guided Flow Matching for Autonomous Driving
  5. 2025nuPlan-R: A Closed-Loop Planning Benchmark for Autonomous Driving via Reactive Multi-Agent Simulation
  6. 2026Adaptive Time Step Flow Matching for Autonomous Driving Motion Planning
  7. 2026ECoSim: Data Efficient Fine-Tuning for Controllable Traffic Simulation
  8. 2026VehicleArena: A Realistic Urban Environment for Multi-Agent Driving