Pith. sign in

REVIEW 6 cited by

Diffusion Dynamics Models with Generative State Estimation for Cloth Manipulation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.11999 v2 pith:TPTH4X36 submitted 2025-03-15 cs.RO cs.CVcs.SYeess.SY

Diffusion Dynamics Models with Generative State Estimation for Cloth Manipulation

classification cs.RO cs.CVcs.SYeess.SY
keywords dynamicsclothmodelsstategenerativeestimationmanipulationmodeling
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Cloth manipulation is challenging due to its highly complex dynamics, near-infinite degrees of freedom, and frequent self-occlusions, which complicate both state estimation and dynamics modeling. Inspired by recent advances in generative models, we hypothesize that these expressive models can effectively capture intricate cloth configurations and deformation patterns from data. Therefore, we propose a diffusion-based generative approach for both perception and dynamics modeling. Specifically, we formulate state estimation as reconstructing full cloth states from partial observations and dynamics modeling as predicting future states given the current state and robot actions. Leveraging a transformer-based diffusion model, our method achieves accurate state reconstruction and reduces long-horizon dynamics prediction errors by an order of magnitude compared to prior approaches. We integrate our dynamics models with model predictive control and show that our framework enables effective cloth folding on real robotic systems, demonstrating the potential of generative models for deformable object manipulation under partial observability and complex dynamics.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. FLASH: Fast Learning via GPU-Accelerated Simulation for High-Fidelity Deformable Manipulation in Minutes

    cs.RO 2026-04 unverdicted novelty 6.0

    A new GPU-accelerated deformable simulation framework trains manipulation policies in minutes using only synthetic data, achieving robust zero-shot transfer to physical robots.

  2. Learn2Fold: Structured Origami Generation with World Model Planning

    cs.GR 2026-02 unverdicted novelty 6.0

    Learn2Fold generates physically valid origami folding sequences from text prompts by decoupling LLM-based program proposals from verification in a learned graph-structured world model.

  3. Scaling Cross-Embodiment World Models for Dexterous Manipulation

    cs.RO 2025-11 conditional novelty 6.0

    A single particle-based world model trained on many simulated robot hands and real human hands can plan dexterous manipulation on robot hands it never trained on.

  4. LaGarNet: Goal-Conditioned Recurrent State-Space Models for Pick-and-Place Garment Flattening

    cs.RO 2025-08 unverdicted novelty 6.0

    The submission's abstract describes a new garment-flattening robot model, but the body is a different paper about document retrieval, making the submission internally inconsistent.

  5. 3D Point World Models: Point Completion Enables More Accurate Dynamics Learning

    cs.RO 2026-06 unverdicted novelty 5.0

    3DPWM completes partial point clouds then learns dynamics on the completed 3D scenes to produce reliable long-horizon rollouts for model-based robotic planning.

  6. Language-Guided Long Horizon Manipulation with LLM-based Planning and Visual Perception

    cs.RO 2025-09 conditional novelty 5.0

    A robot folds cloth from spoken language by decomposing instructions with GPT-4o and grounding each step with a SigLIP2-based pick-and-place perception module.