D2PPO: Diffusion Policy Policy Optimization with Dispersive Loss

• URL: http://arxiv.org/abs/2508.02644v1
• Date: Mon, 04 Aug 2025 17:33:41 GMT
• Title: D2PPO: Diffusion Policy Policy Optimization with Dispersive Loss
• Abstract summary: Diffusion policies excel at robotic manipulation by modeling multimodal action distributions in high-dimensional spaces.<n>Diffusion policies suffer from representation collapse when semantically similar observations are mapped to indistinguishable features.<n>D2PPO introduces dispersive loss regularization that combats representation collapse by treating all hidden representations within each batch as negative pairs.
• Score: 13.132582315951193
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion policies excel at robotic manipulation by naturally modeling multimodal action distributions in high-dimensional spaces. Nevertheless, diffusion policies suffer from diffusion representation collapse: semantically similar observations are mapped to indistinguishable features, ultimately impairing their ability to handle subtle but critical variations required for complex robotic manipulation. To address this problem, we propose D2PPO (Diffusion Policy Policy Optimization with Dispersive Loss). D2PPO introduces dispersive loss regularization that combats representation collapse by treating all hidden representations within each batch as negative pairs. D2PPO compels the network to learn discriminative representations of similar observations, thereby enabling the policy to identify subtle yet crucial differences necessary for precise manipulation. In evaluation, we find that early-layer regularization benefits simple tasks, while late-layer regularization sharply enhances performance on complex manipulation tasks. On RoboMimic benchmarks, D2PPO achieves an average improvement of 22.7% in pre-training and 26.1% after fine-tuning, setting new SOTA results. In comparison with SOTA, results of real-world experiments on a Franka Emika Panda robot show the excitingly high success rate of our method. The superiority of our method is especially evident in complex tasks. Project page: https://guowei-zou.github.io/d2ppo/

Related papers

• OGPO: Sample Efficient Full-Finetuning of Generative Control Policies [53.42266064673132]
  Generative control policies (GCPs) have emerged as effective parameterizations for robot learning.<n>This work introduces Off-policy Generative Policy Optimization (OGPO), a sample-efficient for finetuning GCPs.<n>OGPO achieves state-of-the-art performance on manipulation tasks spanning multi-task settings, high-precision insertion, and dexterous control.
  arXiv  Detail & Related papers  (2026-05-04T18:36:40Z)
• Improving Search Agent with One Line of Code [68.58667107354253]
  Tool-based Agentic Reinforcement Learning (TARL) has emerged as a promising paradigm for training search agents to interact with external tools.<n>We propose textbfSearch textbfAgent textbfPolicy textbfOptimization (textbfSAPO), which stabilizes training via a conditional token-level KL constraint.
  arXiv  Detail & Related papers  (2026-03-10T04:07:39Z)
• SeedPolicy: Horizon Scaling via Self-Evolving Diffusion Policy for Robot Manipulation [50.002839579188624]
  Imitation Learning (IL) enables robots to acquire manipulation skills from expert demonstrations.<n>We propose Self-Evolving Gated Attention (SEGA), a temporal module that maintains a time-evolving latent state via gated attention.<n> integrating SEGA into DP yields Self-Evolving Diffusion Policy (SeedPolicy), which resolves the temporal modeling bottleneck.
  arXiv  Detail & Related papers  (2026-03-05T12:42:53Z)
• SemanticVLA: Semantic-Aligned Sparsification and Enhancement for Efficient Robotic Manipulation [65.6201974979119]
  We propose SemanticVLA, a novel VLA framework that performs Semantic-Hierarchical Sparsification and Enhancement for Efficient Robotic Manipulation.<n>SemanticVLA surpasses OpenVLA on LIBERO benchmark by 21.1% in success rate, while reducing training cost and inference latency by 3.0-fold and 2.7-fold.
  arXiv  Detail & Related papers  (2025-11-13T17:24:37Z)
• Why DPO is a Misspecified Estimator and How to Fix It [17.3051205548111]
  We show that DPO encodes a statistical estimation problem over reward functions induced by a parametric policy class.<n>We propose AuxDPO, which introduces additional auxiliary variables in the DPO loss function to help move towards the RLHF solution.
  arXiv  Detail & Related papers  (2025-10-23T10:30:29Z)
• DM1: MeanFlow with Dispersive Regularization for 1-Step Robotic Manipulation [23.382067451764396]
  Flow-based generative models have emerged as a promising solution to learning distributions of actions.<n>Existing flow-based policies suffer from representation collapse, the inability to distinguish similar visual representations, leading to failures in precise manipulation tasks.<n>We propose DM1, a novel flow matching framework that integrates dispersive regularization into MeanFlow to prevent collapse while maintaining one-step efficiency.
  arXiv  Detail & Related papers  (2025-10-09T07:12:20Z)
• Multimodal Remote Inference [14.609320101695575]
  We study a two-modality scheduling problem that seeks to minimize the ML model's inference error.<n>We show that both modalities share the same threshold and that the index functions and the threshold can be computed efficiently.
  arXiv  Detail & Related papers  (2025-08-11T02:30:44Z)
• Towards Self-Improvement of Diffusion Models via Group Preference Optimization [10.6096255671291]
  Group Preference Optimization (GPO) is an effective self-improvement method that enhances performance without requiring external data.<n>GPO improves the accurate counting and text rendering capabilities of the Stable Diffusion 3.5 Medium by 20 percentage points.<n>As a plug-and-play method, no extra overhead is introduced during inference.
  arXiv  Detail & Related papers  (2025-05-16T10:04:57Z)
• Entropy Controllable Direct Preference Optimization [3.536605202672355]
  We propose a simple modification to DPO, H-DPO, which allows for control over the entropy of the resulting policy.<n>In our experiments, we show that H-DPO outperformed DPO across various tasks, demonstrating superior results in pass@$k$ evaluations for mathematical tasks.
  arXiv  Detail & Related papers  (2024-11-12T07:09:44Z)
• Uncertainty-Penalized Direct Preference Optimization [52.387088396044206]
  We develop a pessimistic framework for DPO by introducing preference uncertainty penalization schemes. The penalization serves as a correction to the loss which attenuates the loss gradient for uncertain samples. We show improved overall performance compared to vanilla DPO, as well as better completions on prompts from high-uncertainty chosen/rejected responses.
  arXiv  Detail & Related papers  (2024-10-26T14:24:37Z)
• Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling [51.38330727868982]
  We show how action chunking impacts the divergence between a learner and a demonstrator.<n>We propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop adaptation.<n>Our method boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks.
  arXiv  Detail & Related papers  (2024-08-30T15:39:34Z)
• D2PO: Discriminator-Guided DPO with Response Evaluation Models [63.71853401569461]
  We propose D2PO, discriminator-guided DPO, for the online setting where preferences are being collected throughout learning. As we collect gold preferences, we use these not only to train our policy, but to train a discriminative response evaluation model to silver-label even more synthetic data for policy training. We show conditions under which silver labeling is most helpful: it is most effective when training the policy with DPO, outperforming traditional PPO, and benefits from maintaining a separate discriminator from the policy model.
  arXiv  Detail & Related papers  (2024-05-02T17:44:41Z)
• Sample Dropout: A Simple yet Effective Variance Reduction Technique in Deep Policy Optimization [18.627233013208834]
  We show that the use of importance sampling could introduce high variance in the objective estimate. We propose a technique called sample dropout to bound the estimation variance by dropping out samples when their ratio deviation is too high.
  arXiv  Detail & Related papers  (2023-02-05T04:44:35Z)
• Robust Policy Optimization in Deep Reinforcement Learning [16.999444076456268]
  In continuous action domains, parameterized distribution of action distribution allows easy control of exploration. In particular, we propose an algorithm called Robust Policy Optimization (RPO), which leverages a perturbed distribution. We evaluated our methods on various continuous control tasks from DeepMind Control, OpenAI Gym, Pybullet, and IsaacGym.
  arXiv  Detail & Related papers  (2022-12-14T22:43:56Z)
• Deterministic and Discriminative Imitation (D2-Imitation): Revisiting Adversarial Imitation for Sample Efficiency [61.03922379081648]
  We propose an off-policy sample efficient approach that requires no adversarial training or min-max optimization. Our empirical results show that D2-Imitation is effective in achieving good sample efficiency, outperforming several off-policy extension approaches of adversarial imitation.
  arXiv  Detail & Related papers  (2021-12-11T19:36:19Z)
• Regressive Domain Adaptation for Unsupervised Keypoint Detection [67.2950306888855]
  Domain adaptation (DA) aims at transferring knowledge from a labeled source domain to an unlabeled target domain. We present a method of regressive domain adaptation (RegDA) for unsupervised keypoint detection. Our method brings large improvement by 8% to 11% in terms of PCK on different datasets.
  arXiv  Detail & Related papers  (2021-03-10T16:45:22Z)
• Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Data [87.61504710345528]
  We propose two strategies for freeing a neural network from tuning with OoD data, while improving its OoD detection performance. We specifically propose to decompose confidence scoring as well as a modified input pre-processing method. Our further analysis on a larger scale image dataset shows that the two types of distribution shifts, specifically semantic shift and non-semantic shift, present a significant difference.
  arXiv  Detail & Related papers  (2020-02-26T04:18:25Z)
• Simple and Effective Prevention of Mode Collapse in Deep One-Class Classification [93.2334223970488]
  We propose two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small.
  arXiv  Detail & Related papers  (2020-01-24T03:44:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.