mlsanityData to 14 Oct 2026

Symphony: Learning Realistic and Diverse Agents for Autonomous Driving Simulation

arXiv 2205.03195 ↗First posted 6 May 2022cs.LG cs.ROcited by 61

Not on the map. Why?

Cited by 24 in End-to-end autonomous driving, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Simulation is a crucial tool for accelerating the development of autonomous vehicles. Making simulation realistic requires models of the human road users who interact with such cars. Such models can be obtained by applying learning from demonstration (LfD) to trajectories observed by cars already on the road. However, existing LfD methods are typically insufficient, yielding policies that frequently collide or drive off the road. To address this problem, we propose Symphony, which greatly improves realism by combining conventional policies with a parallel beam search. The beam search refines these policies on the fly by pruning branches that are unfavourably evaluated by a discriminator. However, it can also harm diversity, i.e., how well the agents cover the entire distribution of realistic behaviour, as pruning can encourage mode collapse. Symphony addresses this issue with a hierarchical approach, factoring agent behaviour into goal generation and goal conditioning. The use of such goals ensures that agent diversity neither disappears during adversarial training nor is pruned away by the beam search. Experiments on both proprietary and open Waymo datasets confirm that Symphony agents learn more realistic and diverse behaviour than several baselines.

Key facts. Cited by 61 papers in the map (data through 14 Oct 2026). Builds on 25 of its 43 references that are in the map. First posted on arXiv 6 May 2022.

Builds on

25 of 43 references.

References · 25

  1. 2022Generative Adversarial Networkscited by 5,627
  2. 2016Generative Adversarial Imitation Learningcited by 1,597
  3. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,546
  4. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  5. 2016End to End Learning for Self-Driving Carscited by 1,009
  6. 2021Offline Reinforcement Learning as One Big Sequence Modeling Problemcited by 644
  7. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 537
  8. 2021NuPlan: A closed-loop ML-based planning benchmark for autonomous vehiclescited by 493
  9. 2019MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Predictioncited by 420
  10. 2018ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worstcited by 397
  11. 2017DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agentscited by 310
  12. 2017Thinking Fast and Slow with Deep Learning and Tree Searchcited by 252
13 more
  1. 2021TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviorscited by 193
  2. 2017DART: Noise Injection for Robust Imitation Learningcited by 191
  3. 2016Value Iteration Networkscited by 160
  4. 2021IntentNet: Learning to Predict Intention from Raw Sensor Datacited by 158
  5. 2021AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehiclescited by 137
  6. 2020SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Drivingcited by 121
  7. 2020Combining Deep Reinforcement Learning and Search for Imperfect-Information Gamescited by 114
  8. 2021SimNet: Learning Reactive Self-driving Simulations from Real-world Observationscited by 77
  9. 2020On the role of planning in model-based deep reinforcement learningcited by 57
  10. 2020Strictly Batch Imitation Learning by Energy-based Distribution Matchingcited by 54
  11. 2018Gated Path Planning Networkscited by 31
  12. 2016Model-based Adversarial Imitation Learningcited by 29
  13. 2018Learning from Demonstration in the Wildcited by 25

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 7; 2023: 20; 2024: 17; 2025: 11; 2026: 6.

61 papers.

Sort:

By area:

  1. 2023UniSim: A Neural Closed-Loop Sensor Simulatorcited by 297
  2. 2023Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Researchcited by 160
  3. 2023MotionLM: Multi-Agent Motion Forecasting as Language Modelingcited by 156
  4. 2024Towards learning-based planning:The nuPlan benchmark for real-world autonomous drivingcited by 104
  5. 2023The Waymo Open Sim Agents Challengecited by 86
  6. 2022BITS: Bi-level Imitation for Traffic Simulationcited by 80
  7. 2025A Survey of World Models for Autonomous Drivingcited by 60
  8. 2022Nocturne: a scalable driving benchmark for bringing multi-agent learning one step closer to the real worldcited by 56
  9. 2023TrafficBots: Towards World Models for Autonomous Driving Simulation and Motion Predictioncited by 56
  10. 2024BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Predictioncited by 50
  11. 2024SceneDiffuser: Efficient and Controllable Driving Simulation Initialization and Rolloutcited by 48
  12. 2023Scenario Diffusion: Controllable Driving Scenario Generation With Diffusioncited by 47
38 more
  1. 2022Hierarchical Model-Based Imitation Learning for Planning in Autonomous Drivingcited by 44
  2. 2024Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Modelscited by 40
  3. 2023Inverse Reinforcement Learning without Reinforcement Learningcited by 36
  4. 2024Hybrid Inverse Reinforcement Learningcited by 32
  5. 2023The Integration of Prediction and Planning in Deep Learning Automated Driving Systems: A Reviewcited by 29
  6. 2023Data-driven Traffic Simulation: A Comprehensive Reviewcited by 29
  7. 2025Scenario Dreamer: Vectorized Latent Diffusion for Generating Driving Simulation Environmentscited by 28
  8. 2024CtRL-Sim: Reactive and Controllable Driving Agents with Offline Reinforcement Learningcited by 27
  9. 2023Multiverse Transformer: 1st Place Solution for Waymo Open Sim Agents Challenge 2023cited by 25
  10. 2023Rethinking Imitation-based Planner for Autonomous Drivingcited by 25
  11. 2024AdvDiffuser: Generating Adversarial Safety-Critical Driving Scenarios via Guided Diffusioncited by 20
  12. 2025S4-Driver: Scalable Self-Supervised Driving Multimodal Large Language Modelwith Spatio-Temporal Visual Representationcited by 19
  13. 2023What Truly Matters in Trajectory Prediction for Autonomous Driving?cited by 17
  14. 2024Human-compatible driving partners through data-regularized self-play reinforcement learningcited by 17
  15. 2022Critic Sequential Monte Carlocited by 16
  16. 2025UniMM: A Unified Mixture Model Framework for Multi-Agent Simulationcited by 13
  17. 2025Generating Multimodal Driving Scenes via Next-Scene Predictioncited by 12
  18. 2024Learning to Drive via Asymmetric Self-Playcited by 11
  19. 2023Editing Driver Character: Socially-Controllable Behavior Generation for Interactive Traffic Simulationcited by 10
  20. 2023Categorical Traffic Transformer: Interpretable and Diverse Behavior Prediction with Tokenized Latentcited by 10
  21. 2024Non-Adversarial Inverse Reinforcement Learning via Successor Feature Matchingcited by 10
  22. 2022Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domainscited by 7
  23. 2024Reward-free World Models for Online Imitation Learningcited by 7
  24. 2023Resolving uncertainty on the fly: Modeling adaptive driving behavior as active inferencecited by 6
  25. 2024TorchDriveEnv: A Reinforcement Learning Benchmark for Autonomous Driving with Reactive, Realistic, and Diverse Non-Playable Characterscited by 6
  26. 2022Learning from Demonstrations of Critical Driving Behaviours Using Driver's Risk Fieldcited by 4
  27. 2024LoRD: Adapting Differentiable Driving Policies to Distribution Shiftscited by 4
  28. 2025Safety-Critical Traffic Simulation with Adversarial Transfer of Driving Intentionscited by 4
  29. 2023A Counterfactual Safety Margin Perspective on the Scoring of Autonomous Vehicles' Riskinesscited by 3
  30. 2023Hierarchical Imitation Learning for Stochastic Environmentscited by 3
  31. 2023Generating Transferable Adversarial Simulation Scenarios for Self-Driving via Neural Renderingcited by 3
  32. 2024Adversarial and Reactive Traffic Entities for Behavior-Realistic Driving Simulation: A Reviewcited by 3
  33. 2023Learning An Active Inference Model of Driver Perception and Control: Application to Vehicle Car-Followingcited by 2
  34. 2026VectorWorld: Efficient Streaming World Model via Diffusion Flow on Vector Graphscited by 2
  35. 2026RLFTSim: Realistic and Controllable Multi-Agent Traffic Simulation via Reinforcement Learning Fine-Tuningcited by 2
  36. 2022P4P: Conflict-Aware Motion Prediction for Planning in Autonomous Drivingcited by 1
  37. 2023On Learning the Tail Quantiles of Driving Behavior Distributions via Quantile Regression and Flowscited by 1
  38. 2024LitSim: A Conflict-aware Policy for Long-term Interactive Traffic Simulationcited by 1

50 most cited of 61 All 61 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2018Multi-Agent Imitation Learning for Driving Simulation
  2. 2026TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale
  3. 2026AutoWorld: Learning Multi-Agent Traffic Simulation with Self-Supervised World Models
  4. 2025Autonomous Vehicle Path Planning by Searching With Differentiable Simulation
  5. 2022Driver Dojo: A Benchmark for Generalizable Reinforcement Learning for Autonomous Driving
  6. 2025V-Max: A Reinforcement Learning Framework for Autonomous Driving
  7. 2023Multi-Agent Deep Reinforcement Learning for Cooperative and Competitive Autonomous Vehicles using AutoDRIVE Ecosystem
  8. 2026Scaling Curriculum Learning For Autonomous Driving