Frequency-Guided Action Diffusion via Sub-Frequency Manifold Traversal
Summary
This paper introduces the Frequency Guidance Operator (FGO), a method for diffusion policies that smooths action generation by steering noisy samples through intermediate sub-frequency manifolds, improving performance on robotic manipulation tasks.
View Cached Full Text
Cached at: 06/01/26, 07:18 AM
Paper page - Frequency-Guided Action Diffusion via Sub-Frequency Manifold Traversal
Source: https://huggingface.co/papers/2605.27919
Abstract
Frequency Guidance Operator enables smooth action generation in diffusion policies by steering noisy samples through intermediate sub-frequency manifolds, improving robotic manipulation performance.
Learning visuomotor policies viabehavior cloningtypically involves mimicking expert demonstrations collected by human operators. However, natural human demonstrations inherently containhigh-frequency noise, such as intermittent jerks, pauses, and action jitter. Training policies to directly imitate these raw trajectories inevitably causes the model to inherit these suboptimal behaviors. This pathology is particularly pronounced in diffusion-based policies, where iterativedenoising stepscan inadvertently amplify high-frequency artifacts at the expense of meaningful fine-grained details. To address these limitations, we present a novel frequency-based algorithm that enables implicitspectral maneuveringand smooth action generation. Our method,Frequency Guidance Operator(FGO), steers the generation process of diffusion polices by progressively driving the noisy samples through intermediate sub-frequency manifolds with expanding spectral bands. Validated on 15robotic manipulationtasks from 5 benchmarks, FGO achieves superior performance in enhancingaction smoothnessandtemporal consistencywhile preserving the details necessary for successful task execution. Project website: https://henrywjl.github.io/frequency-guidance-operator/
View arXiv pageView PDFProject pageGitHub0Add to collection
Get this paper in your agent:
hf papers read 2605\.27919
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2605.27919 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2605.27919 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2605.27919 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Spectral Guidance for Flexible and Efficient Control of Diffusion Models
Introduces Spectral Guidance, a framework for controlling diffusion models by leveraging low-dimensional representations of the diffusion process, enabling flexible and stable control without task-specific retraining or backpropagation through the denoiser.
Learning Foresight without Explicit Trajectories for 3D Diffusion Policies
Introduces Movement Trend Guidance to enhance 3D diffusion policies in robotic manipulation by providing foresight without explicit trajectories, achieving improved performance on benchmarks like RoboTwin2.0 and LIBERO-40.
Self-Augmented Diffusion Guidance for Physics-Informed Generation
The paper proposes a self-augmented diffusion guidance method to incorporate physical laws into diffusion models, reducing deviations from true dynamics and enabling faster generation.
Class-frequency Guided Noise Schedule for Diffusion Models
This paper proposes a class-frequency guided noise schedule for diffusion models that assigns larger-scale noises to low-frequency classes to improve generation quality on imbalanced datasets, demonstrating substantial improvements over baselines.
A Frozen Pixel-Space Diffusion Model Can Guide Itself with Its Own Samples
This paper introduces Synthetic Self-Guidance (SSG), a method that attaches a lightweight prediction head to a frozen pretrained pixel-space diffusion model, using the discrepancy between intermediate and final predictions as self-guidance during sampling. It shows that model-generated samples suffice for training the head, improving FID by over 50% on several variants without classifier-free guidance and enhancing strong baselines with CFG.