mlsanityData to 7 Oct 2026

Unpacking the Individual Components of Diffusion Policy

arXiv 2412.00084 ↗First posted 27 Nov 2024cs.LG cs.ROcited by 0

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

also in 1 more
Abstract (arXiv)

Imitation Learning presents a promising approach for learning generalizable and complex robotic skills. The recently proposed Diffusion Policy generates robot action sequences through a conditional denoising diffusion process, achieving state-of-the-art performance compared to other imitation learning methods. This paper summarizes five key components of Diffusion Policy: 1) observation sequence input; 2) action sequence execution; 3) receding horizon; 4) U-Net or Transformer network architecture; and 5) FiLM conditioning. By conducting experiments across ManiSkill and Adroit benchmarks, this study aims to elucidate the contribution of each component to the success of Diffusion Policy in various scenarios. We hope our findings will provide valuable insights for the application of Diffusion Policy in future research and industry.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 30 of its 33 references that are in the map. First posted on arXiv 27 Nov 2024.

Builds on

30 of 33 references.

Closest ancestors · 9

Same direction, newest first

  1. 2024Diffusion Policy Policy Optimization1 Sep
  2. 2024Diffusion-DICE: In-Sample Diffusion Guidance for Offline Reinforcement Learning29 Jul
  3. 20243D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations6 Mar
  4. 2023Consistency Models as a Rich and Efficient Policy Class for Reinforcement Learning29 Sep
  5. 2023Policy Representation via Diffusion Probability Model for Reinforcement Learning22 May
  6. 2023IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies20 Apr
  7. 2023Goal-Conditioned Imitation Learning using Score-based Diffusion Policies5 Apr
  8. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  9. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning12 Aug

Other references · 21

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2022Scalable Diffusion Models with Transformerscited by 5,931
  6. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  7. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  8. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 699
  9. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 685
  10. 2021Implicit Behavioral Cloningcited by 442
  11. 2023ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skillscited by 313
  12. 2022Behavior Transformers: Cloning k modes with one stonecited by 299
9 more
  1. 2021DexMV: Imitation Learning for Dexterous Manipulation from Human Videoscited by 194
  2. 2021ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrationscited by 193
  3. 2020Language-Conditioned Imitation Learning for Robot Manipulation Taskscited by 136
  4. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstrationcited by 92
  5. 2020State-Only Imitation Learning for Dexterous Manipulationcited by 90
  6. 2021Error-Aware Imitation Learning from Teleoperation Data for Mobile Manipulationcited by 59
  7. 2020Deep Imitation Learning for Bimanual Robotic Manipulationcited by 49
  8. 2023LocoMuJoCo: A Comprehensive Imitation Learning Benchmark for Locomotioncited by 31
  9. 2022Bayesian Imitation Learning for End-to-End Mobile Manipulationcited by 13

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025CCDP: Composition of Conditional Diffusion Policies with Guided Sampling
  2. 2024Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Models
  3. 2025Two-Steps Diffusion Policy for Robotic Manipulation via Genetic Denoising
  4. 2024Prediction with Action: Visual Policy Learning via Joint Denoising Process
  5. 2025Efficient Task-specific Conditional Diffusion Policies: Shortcut Model Acceleration and SO(3) Optimization
  6. 2025Time-Unified Diffusion Policy with Action Discrimination for Robotic Manipulation
  7. 2024Hierarchical Diffusion Policy: manipulation trajectory generation via contact guidance
  8. 2025Demystifying Diffusion Policies: Action Memorization and Simple Lookup Table Alternatives