mlsanityData to 14 Oct 2026

Discrete Contrastive Learning for Diffusion Policies in Autonomous Driving

arXiv 2503.05229 ↗First posted 7 Mar 2025cs.RO cs.AIcited by 1

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

Abstract (arXiv)

Learning to perform accurate and rich simulations of human driving behaviors from data for autonomous vehicle testing remains challenging due to human driving styles' high diversity and variance. We address this challenge by proposing a novel approach that leverages contrastive learning to extract a dictionary of driving styles from pre-existing human driving data. We discretize these styles with quantization, and the styles are used to learn a conditional diffusion policy for simulating human drivers. Our empirical evaluation confirms that the behaviors generated by our approach are both safer and more human-like than those of the machine-learning-based baseline methods. We believe this has the potential to enable higher realism and more effective techniques for evaluating and improving the performance of autonomous vehicles.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 44 of its 70 references that are in the map. First posted on arXiv 7 Mar 2025.

Builds on

44 of 70 references.

Closest ancestors · 9

Same direction, newest first

  1. 2023Generative Skill Chaining: Long-Horizon Skill Planning with Diffusion Models13 Oct
  2. 2023Consistency Models as a Rich and Efficient Policy Class for Reinforcement Learning29 Sep
  3. 2023Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning29 May
  4. 2023IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies20 Apr
  5. 2023Goal-Conditioned Imitation Learning using Score-based Diffusion Policies5 Apr
  6. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  7. 2023Imitating Human Behaviour with Diffusion Models25 Jan
  8. 2022Is Conditional Generative Modeling all you need for Decision-Making?28 Nov
  9. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning12 Aug

Other references · 35

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  3. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  5. 2017Neural Discrete Representation Learningcited by 4,712
  6. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  7. 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,501
  8. 2022Video Diffusion Modelscited by 2,305
  9. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  10. 2023Consistency Modelscited by 1,822
  11. 2016Generative Adversarial Imitation Learningcited by 1,597
  12. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,354
23 more
  1. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,061
  2. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 640
  3. 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 639
  4. 2018An Algorithmic Perspective on Imitation Learningcited by 459
  5. 2021Implicit Behavioral Cloningcited by 452
  6. 2023Finite Scalar Quantization: VQ-VAE Made Simplecited by 440
  7. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 434
  8. 2022Behavior Transformers: Cloning k modes with one stonecited by 304
  9. 2022Flexible Diffusion Modeling of Long Videoscited by 294
  10. 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 200
  11. 2019A Divergence Minimization Perspective on Imitation Learning Methodscited by 190
  12. 2023Noise2Music: Text-conditioned Music Generation with Diffusion Modelscited by 173
  13. 2020Contrastive Learning of General-Purpose Audio Representationscited by 169
  14. 2019Imitation Learning as f-Divergence Minimizationcited by 123
  15. 2017Imitating Driver Behavior with Generative Adversarial Networkscited by 93
  16. 2021Imitation Learning: Progress, Taxonomies and Challengescited by 58
  17. 2023TACO: Temporal Latent Action-Driven Contrastive Loss for Visual Reinforcement Learningcited by 57
  18. 2020Robust Training of Vector Quantized Bottleneck Modelscited by 54
  19. 2021Long-Short Temporal Contrastive Learning of Video Transformerscited by 42
  20. 2024Deep Generative Models in Robotics: A Survey on Learning from Multimodal Demonstrationscited by 35
  21. 2023Data-driven Traffic Simulation: A Comprehensive Reviewcited by 29
  22. 2017The Value of Inferring the Internal State of Traffic Participants for Autonomous Freeway Drivingcited by 15
  23. 2022Modeling Human Behavior Part I -- Learning and Belief Approachescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026CoMI-IRL: Contrastive Multi-Intention Inverse Reinforcement Learningcited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2021Towards Learning Generalizable Driving Policies from Restricted Latent Representations
  2. 2021Urban Driver: Learning to Drive from Real-world Demonstrations Using Policy Gradients
  3. 2026Aligning Microscopic Vehicle and Macroscopic Traffic Statistics: Reconstructing Driving Behavior from Partial Data
  4. 2025Learning Personalized Driving Styles via Reinforcement Learning from Human Feedback
  5. 2019Urban Driving with Conditional Imitation Learning
  6. 2025Controllable Latent Diffusion for Traffic Simulation
  7. 2025LAP: Fast LAtent Diffusion Planner for Autonomous Driving
  8. 2026STAGE: STyle-controllable Action GEneration for personalized autonomous driving