Hugging Face Daily PapersCan Li, Jie Gu, Zishun DengSep 17, 20261 min readpaperadvanced

DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation

Summary

DeformSmith is a framework that generates physically plausible deformable assets for robot manipulation from a text prompt or a single image, using a hierarchical construction process guided by a shared physics harness. It outperforms prior baselines in visual fidelity and physical realism while also producing interaction data for downstream tasks.

Hierarchical agentic construction iteratively builds geometry, material models, and interaction behaviors, refining each stage with physics simulation feedback.
A shared physics‑grounded harness ties together visual generation and physical property estimation, ensuring consistency between appearance and dynamics.
The system can start from minimal input (text or one image) and produce assets ready for simulation and real‑world manipulation pipelines.
Quantitative evaluation shows higher visual quality and physical plausibility than PhysGen3D, PhysGM, and PhysX‑Omni baselines.

Engineers building simulation environments for robotic manipulation need realistic deformable models; this work offers an automated way to create them from sparse inputs.

7/10

robotics generative models simulation physics deformable objects

Related reading

SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image arxiv.org
The paper presents SNAP3D, a physics‑guided pipeline that converts a single image into a set of 3D parts that can be assembled without interpenetration. By using simulation‑driven connector placement and a new physics‑based evaluation, the method yields assemblies that are both geometrically accurate and stable enough for 3D printing.
Hugging Face Daily Papersarxiv.orgYu-Rou Tuan, Hao-Tang Tsui, Nicolas Ugrinovic11d1 minpaper
HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness arxiv.org
HarnessVLN introduces a zero‑shot, training‑free embodied navigation framework that wraps a multimodal LLM in an "Agent Harness" – a tool‑based protocol that validates planner actions against spatial evidence, tracks progress with hierarchical event memory, and maintains a persistent spatiotemporal graph for recovery. The system works for instruction‑following and object‑goal tasks, achieving 60.…
Hugging Face Daily Papersarxiv.orgYang Chen, Lirong Che, Zhenyu Huang8d1 minpaper
ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement arxiv.org
ModularRSI proposes a contrastive, modular approach to evolve agent harnesses by independently updating five functional modules and aggregating evidence across 2,000 benchmark‑disjoint tasks. Experiments on TB2.0 and SWE‑Bench Verified show consistent gains on unseen tasks and cross‑model transfer.
Hugging Face Daily Papersarxiv.orgSiwei Wu, Jincheng Ren, Yizhi Li8d1 minpaper
PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Control arxiv.org
PhysStream introduces a two‑stage autoregressive video generator that uses online‑derived positional and tracking maps (structured scene memory) and sparse velocity‑increment signals to enable fine‑grained, physics‑grounded control of multi‑object tabletop scenes. It cuts motion distribution error by 33 % and trajectory error by 12 % versus strong baselines, and wins 85 % of human preference test…
Hugging Face Daily Papersarxiv.orgChuhao Chen, Peter Wonka, Chaoyang Wang7d1 minpaper
FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observations arxiv.org
FAMOS is a feed-forward model that predicts 3D articulation, including movable-part segmentation and joint parameters, from sparse, unordered point clouds. It leverages multiple observations through a Multi-state Articulation Transformer and a novel observed articulation span objective, outperforming existing baselines.
Hugging Face Daily Papersarxiv.orgKevin Qu, Tao Sun, Massimiliano Viola5d1 minpaper
Fingers as Legs: Learning Self-Supported Locomotion and Manipulation with an Anthropomorphic Hand arxiv.org
The authors train an anthropomorphic robotic hand to crawl, steer, and recover from falls using its fingers for both support and manipulation, via a reinforcement‑learning reward formulation tuned to the hand's asymmetry. Sim‑to‑real experiments show faster locomotion than quadruped‑style rewards and successful untethered tasks without onboard vision.
Hugging Face Daily Papersarxiv.orgAmirhossein Kazemipour, Hehui Zheng, Robert Katzschmann7d1 minpaper

DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation

Summary

Related reading

SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Imagearxiv.org

HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harnessarxiv.org

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvementarxiv.org

PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Controlarxiv.org

FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observationsarxiv.org

Fingers as Legs: Learning Self-Supported Locomotion and Manipulation with an Anthropomorphic Handarxiv.org

SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image arxiv.org

HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness arxiv.org

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement arxiv.org

PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Control arxiv.org

FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observations arxiv.org

Fingers as Legs: Learning Self-Supported Locomotion and Manipulation with an Anthropomorphic Hand arxiv.org