# MoE-DP: An MoE-Enhanced Diffusion Policy for Robust Long-Horizon Robotic Manipulation with Skill Decomposition and Failure Recovery

arXiv: https://arxiv.org/abs/2511.05007 · HTML: https://mlsanity.com/paper/2511.05007/

## Key facts

- Cited by 10 papers 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 39 of its 43 references that are in the map.
- First posted on arXiv 7 Nov 2025.

## Placed in

- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)

## Builds on (most cited first, 30 of 39)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Planning with Diffusion for Flexible Behavior Synthesis](https://mlsanity.com/paper/2205.09991/) (2022)
- [DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models](https://mlsanity.com/paper/2401.06066/) (2024)
- [Is Conditional Generative Modeling all you need for Decision-Making?](https://mlsanity.com/paper/2211.15657/) (2022)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations](https://mlsanity.com/paper/2310.17596/) (2023)
- [A Survey on Mixture of Experts in Large Language Models](https://mlsanity.com/paper/2407.06204/) (2024)
- [Goal-Conditioned Imitation Learning using Score-based Diffusion Policies](https://mlsanity.com/paper/2304.02532/) (2023)
- [Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts](https://mlsanity.com/paper/2408.15664/) (2024)
- [Generative Modelling With Inverse Heat Dissipation](https://mlsanity.com/paper/2206.13397/) (2022)
- [Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Review](https://mlsanity.com/paper/2501.09685/) (2025)
- [Bottom-Up Skill Discovery from Unsegmented Demonstrations for Long-Horizon Robot Manipulation](https://mlsanity.com/paper/2109.13841/) (2021)
- [XSkill: Cross Embodiment Skill Discovery](https://mlsanity.com/paper/2307.09955/) (2023)
- [SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution](https://mlsanity.com/paper/2312.11598/) (2023)
- [Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learning](https://mlsanity.com/paper/2412.12953/) (2024)
- [Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysis](https://mlsanity.com/paper/2502.00935/) (2025)
- [DynaGuide: Steering Diffusion Polices with Active Dynamic Guidance](https://mlsanity.com/paper/2506.13922/) (2025)
- [Inference-Time Policy Steering through Human Interactions](https://mlsanity.com/paper/2411.16627/) (2024)
- [CLONE: Closed-Loop Whole-Body Humanoid Teleoperation for Long-Horizon Tasks](https://mlsanity.com/paper/2506.08931/) (2025)
- [Multi-Task Interactive Robot Fleet Learning with Visual World Models](https://mlsanity.com/paper/2410.22689/) (2024)
- [Model Based Planning with Energy Based Models](https://mlsanity.com/paper/1909.06878/) (2019)

## Built on by (most cited first, 10 of 10)

- [VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model](https://mlsanity.com/paper/2602.12063/) (2026)
- [SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation](https://mlsanity.com/paper/2602.16863/) (2026)
- [GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization](https://mlsanity.com/paper/2605.12369/) (2026)
- [Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning](https://mlsanity.com/paper/2604.16683/) (2026)
- [Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training](https://mlsanity.com/paper/2604.23121/) (2026)
- [SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation](https://mlsanity.com/paper/2606.10305/) (2026)
- [EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents](https://mlsanity.com/paper/2609.01281/) (2026)
- [MoRI: Mixture of RL and IL Experts for Long-Horizon Manipulation Tasks](https://mlsanity.com/paper/2604.10165/) (2026)
- [Move-Then-Operate: Behavioral Phasing for Human-Like Robotic Manipulation](https://mlsanity.com/paper/2604.23620/) (2026)
- [Factor-Aware Mixture-of-Experts with Pretrained Encoder for Combinatorial Generalization](https://mlsanity.com/paper/2606.21100/) (2026)

## Abstract (arXiv)

Diffusion policies have emerged as a powerful framework for robotic visuomotor control, yet they often lack the robustness to recover from subtask failures in long-horizon, multi-stage tasks and their learned representations of observations are often difficult to interpret. In this work, we propose the Mixture of Experts-Enhanced Diffusion Policy (MoE-DP), where the core idea is to insert a Mixture of Experts (MoE) layer between the visual encoder and the diffusion model. This layer decomposes the policy's knowledge into a set of specialized experts, which are dynamically activated to handle different phases of a task. We demonstrate through extensive experiments that MoE-DP exhibits a strong capability to recover from disturbances, significantly outperforming standard baselines in robustness. On a suite of 6 long-horizon simulation tasks, this leads to a 36% average relative improvement in success rate under disturbed conditions. This enhanced robustness is further validated in the real world, where MoE-DP also shows significant performance gains. We further show that MoE-DP learns an interpretable skill decomposition, where distinct experts correspond to semantic task primitives (e.g., approaching, grasping). This learned structure can be leveraged for inference-time control, allowing for the rearrangement of subtasks without any re-training.Our video and code are available at the https://moe-dp-website.github.io/MoE-DP-Website/.