PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration
PGDG is a novel data generation framework that expands a single demonstration into a diverse dataset of physically plausible recovery behaviors for bimanual manipulation, without human labeling. It iteratively improves data quality using a physics-grounded sampler and curator, significantly outperforming spatial-only augmentation on both simulated and real-world tasks, and enables fine-tuning of foundation models like GR00T.
[2605.21710] PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration
[Submitted on 20 May 2026]
Title:PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration
View a PDF of the paper titled PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration, by Cunxi Dai and 6 other authors
View PDF HTML (experimental)
Abstract:Behavior cloning for contact-rich bimanual manipulation remains challenging because diverse demonstrations are expensive to collect, and even small disturbances can push the system into off-manifold states where no recovery supervision is available. We propose PGDG, a data generation framework with zero-shot curation that expands a single demonstration into a compact dataset of physically plausible, successful, and diverse recovery behaviors without additional human labeling. PGDG iterates between a physics-grounded sampler and a dataset curator, where the curator selects informative, non-redundant, and recoverable behaviors to update the sampling distribution toward under-covered recovery modes, and the sampler draws physically plausible rollout candidates from this updated distribution and retains successful trajectories. To further improve data quality, PGDG applies short-horizon sampling-based control to relabel selected risky states with corrective actions. Across four bimanual manipulation tasks, PGDG consistently outperforms spatial-only augmentation in both simulation and zero-shot real-world transfer. On RotateBox-Pitch, success improves from 38% to 93% in simulation and from 35% to 82% in the real world. PGDG also enables effective foundation models fine-tuning such as GR00T, increasing success from 46% to 77%. Additional results are available in our website: this https URL.
Subjects:
Robotics (cs.RO)
Cite as: arXiv:2605.21710 [cs.RO]
(or arXiv:2605.21710v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2605.21710
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Cunxi Dai [view email] [v1] Wed, 20 May 2026 20:14:24 UTC (1,844 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration, by Cunxi Dai and 6 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.RO
new | recent | 2026-05
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)