mlsanityData to 7 Oct 2026

Crossway Diffusion: Improving Diffusion-based Visuomotor Policy via Self-supervised Learning

arXiv 2307.01849 ↗First posted 4 Jul 2023cs.RO cs.CV cs.LGcited by 37

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

also in 1 more
Abstract (arXiv)

Sequence modeling approaches have shown promising results in robot imitation learning. Recently, diffusion models have been adopted for behavioral cloning in a sequence modeling fashion, benefiting from their exceptional capabilities in modeling complex data distributions. The standard diffusion-based policy iteratively generates action sequences from random noise conditioned on the input states. Nonetheless, the model for diffusion policy can be further improved in terms of visual representations. In this work, we propose Crossway Diffusion, a simple yet effective method to enhance diffusion-based visuomotor policy learning via a carefully designed state decoder and an auxiliary self-supervised learning (SSL) objective. The state decoder reconstructs raw image pixels and other state information from the intermediate representations of the reverse diffusion process. The whole model is jointly optimized by the SSL objective and the original diffusion loss. Our experiments demonstrate the effectiveness of Crossway Diffusion in various simulated and real-world robot tasks, confirming its consistent advantages over the standard diffusion-based policy and substantial improvements over the baselines.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 71 of its 96 references that are in the map. First posted on arXiv 4 Jul 2023.

Builds on

71 of 96 references.

Closest ancestors · 5

Same direction, newest first

  1. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  2. 2023Diffusion Model-Augmented Behavioral Cloning26 Feb
  3. 2023Imitating Human Behaviour with Diffusion Models25 Jan
  4. 2022Is Conditional Generative Modeling all you need for Decision-Making?28 Nov
  5. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning12 Aug

Other references · 66

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  4. 2020Language Models are Few-Shot Learnerscited by 32,199
  5. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  6. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  7. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  8. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  9. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  10. 2020Denoising Diffusion Implicit Modelscited by 9,655
  11. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  12. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
38 more
  1. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  4. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  5. 2023Consistency Modelscited by 1,778
  6. 2016Generative Adversarial Imitation Learningcited by 1,576
  7. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  8. 2018World Modelscited by 1,444
  9. 2022VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Trainingcited by 1,305
  10. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
  11. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  12. 2021A Minimalist Approach to Offline Reinforcement Learningcited by 882
  13. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 773
  14. 2018Recurrent World Models Facilitate Policy Evolutioncited by 761
  15. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
  16. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 699
  17. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 685
  18. 2021Offline Reinforcement Learning as One Big Sequence Modeling Problemcited by 636
  19. 2023VideoMAE V2: Scaling Video Masked Autoencoders with Dual Maskingcited by 516
  20. 2023Learning Universal Policies via Text-Guided Video Generationcited by 506
  21. 2018An Algorithmic Perspective on Imitation Learningcited by 455
  22. 2017Time-Contrastive Networks: Self-Supervised Learning from Videocited by 453
  23. 2021Implicit Behavioral Cloningcited by 442
  24. 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 424
  25. 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 286
  26. 2019Improving Sample Efficiency in Model-Free Reinforcement Learning from Imagescited by 268
  27. 2020Decoupling Representation Learning from Reinforcement Learningcited by 263
  28. 2022Masked Visual Pre-training for Motor Controlcited by 246
  29. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 228
  30. 2022Online Decision Transformercited by 194
  31. 2023Scaling Robot Learning with Semantically Imagined Experiencecited by 175
  32. 2018Deep Variational Reinforcement Learning for POMDPscited by 122
  33. 2019Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEscited by 116
  34. 2019Unsupervised Learning of Object Keypoints for Perception and Controlcited by 109
  35. 2019Self-Supervised Correspondence in Visuomotor Policy Learningcited by 106
  36. 2020Bootstrap Latent-Predictive Representations for Multitask Reinforcement Learningcited by 101
  37. 2017Vision-Based Multi-Task Manipulation for Inexpensive Robots Using End-To-End Learning from Demonstrationcited by 100
  38. 2023Synthetic Experience Replaycited by 100

50 most cited of 66 All 66 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 13; 2025: 13; 2026: 8.

37 papers.

Sort:

By area:

  1. 2025DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledgecited by 203
  2. 2024Multimodal Diffusion Transformer: Learning Versatile Behavior from Multimodal Goalscited by 131
  3. 2024Diffusion-based Reinforcement Learning via Q-weighted Variational Policy Optimizationcited by 99
  4. 2023Diffusion Models for Reinforcement Learning: A Surveycited by 84
  5. 2024Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Modelscited by 51
  6. 2024LLaRA: Supercharging Robot Learning Data for Vision-Language Policycited by 51
  7. 2024Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learningcited by 46
  8. 2025NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidancecited by 46
  9. 2025Diffusion Models for Robotic Manipulation: A Surveycited by 39
  10. 2023Boosting Continuous Control with Consistency Policycited by 36
  11. 2024Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learningcited by 34
  12. 2024Generative Image as Action Modelscited by 18
25 more
  1. 2023Limited Data, Unlimited Potential: A Study on ViTs Augmented by Masked Autoencoderscited by 17
  2. 2024Diffusion Policy Attacker: Crafting Adversarial Attacks for Diffusion-based Policiescited by 17
  3. 2025GenPO: Generative Diffusion Models Meet On-Policy Reinforcement Learningcited by 16
  4. 2025Revisiting Multi-Agent World Modeling from a Diffusion-Inspired Perspectivecited by 9
  5. 2024DISCO: Language-Guided Manipulation with Diffusion Policies and Constrained Inpaintingcited by 8
  6. 2025CDP: Towards Robust Autoregressive Visuomotor Policy Learning via Causal Diffusioncited by 8
  7. 2025Pixel Motion Diffusion is What We Need for Robot Controlcited by 7
  8. 2026One-Step Flow Policy: Self-Distillation for Fast Visuomotor Policiescited by 5
  9. 2025Adversarial Diffusion for Robust Reinforcement Learningcited by 4
  10. 2026SeedPolicy: Horizon Scaling via Self-Evolving Diffusion Policy for Robot Manipulationcited by 4
  11. 2024Fibottention: Inceptive Visual Representation Learning with Diverse Attention Across Headscited by 3
  12. 2025Robot Motion Planning using One-Step Diffusion with Noise-Optimized Approximate Motionscited by 3
  13. 2024Task-Agnostic Pre-training and Task-Guided Fine-tuning for Versatile Diffusion Plannercited by 1
  14. 2025Controllable Latent Diffusion for Traffic Simulationcited by 1
  15. 2024Learning Generalizable 3D Manipulation With 10 Demonstrationscited by 0
  16. 2024Hierarchical Diffusion Policy: manipulation trajectory generation via contact guidancecited by 0
  17. 2025CoStoDet-DDPM: Collaborative Training of Stochastic and Deterministic Models Improves Surgical Workflow Anticipation and Recognitioncited by 0
  18. 2025Grounding Bodily Awareness in Visual Representations for Efficient Policy Learningcited by 0
  19. 2025SeFA-Policy: Fast and Accurate Visuomotor Policy Learning with Selective Flow Alignmentcited by 0
  20. 2026GraspLDP: Towards Generalizable Grasping Policy via Latent Diffusioncited by 0
  21. 2026Advantage-Guided Diffusion for Model-Based Reinforcement Learningcited by 0
  22. 2026From Denoising to Decision Making: A Survey on Diffusion Model-Enabled Deep Reinforcement Learning for Wireless Networkscited by 0
  23. 2026Off the Rails: Hijacking the Scoring Head in Generative End-to-End Driving Planners with Safety-Violating Adversarial Perturbationscited by 0
  24. 2026A Single Diffusion-Policy Controller for Multi-Task Block Pushing with Zero-Shot Sim-to-Real Transfercited by 0
  25. 2026Loop-Free Inverse Reinforcement Learning via Sequential Value Recovery with Q-Score Matchingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025DINOv3-Diffusion Policy: Self-Supervised Large Visual Model for Visuomotor Diffusion Policy Learning
  2. 2024One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillation
  3. 2025Sample from What You See: Visuomotor Policy Learning via Diffusion Bridge with Observation-Embedded Stochastic Differential Equation
  4. 2026Latent Diffusion Policy: Shaping Latent Spaces for Diffusion-Based Robotic Manipulation
  5. 2025Spatial-Temporal Aware Visuomotor Diffusion Policy Learning
  6. 2026Hyper-DP3: Frequency-Aware Right-Sizing of 3D Diffusion Policies for Visuomotor Control
  7. 2025H3DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning
  8. 2025Latent Diffusion Planning for Imitation Learning