{"id":1177606,"date":"2026-07-02T08:38:09","date_gmt":"2026-07-02T15:38:09","guid":{"rendered":"https:\/\/find.codeghost.online\/en-us\/research\/publication\/deformgen-dynamics-based-topology-augmentation-for-deformable-manipulation-policy-learning\/"},"modified":"2026-07-10T15:24:54","modified_gmt":"2026-07-10T22:24:54","slug":"deformgen-dynamics-based-topology-augmentation-for-deformable-manipulation-policy-learning","status":"publish","type":"msr-research-item","link":"https:\/\/find.codeghost.online\/en-us\/research\/publication\/deformgen-dynamics-based-topology-augmentation-for-deformable-manipulation-policy-learning\/","title":{"rendered":"DeformGen: Dynamics-Based Topology Augmentation for Deformable Manipulation Policy Learning"},"content":{"rendered":"\n\n\n<p class=\"wp-block-paragraph\">Demonstration augmentation is proposed for cost-efficient data acquisition, but existing methods are fundamentally limited in deformable manipulation due to two challenges: (1) the state space is high-dimensional with physics-induced constraints, making valid configurations impossible to reach via low-dimensional pose perturbations; and (2) trajectory transfer is non-equivariant, as material points no longer move rigidly together under deformation. We present DeformGen, a dynamics-based augmentation framework that achieves topological diversity for deformable objects. For the state challenge, DeformGen expands the valid state distribution by applying localized physical disturbances and forward-simulating the dynamics to obtain topology-coherent, physically plausible deformable states. For the trajectory challenge, DeformGen transfers source manipulation trajectories via deformation-field warping, which lifts per-particle displacements into a continuous spatial function to adapt the end-effector trajectory consistently with the deformed geometry. In this way, our method jointly augments the state distribution and its associated manipulation behavior. Experiments on high-fidelity deformable manipulation benchmarks show that DeformGen generally improves policy learning compared with training on the original demonstrations alone and with rigid-style augmentation baselines.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Demonstration augmentation is proposed for cost-efficient data acquisition, but existing methods are fundamentally limited in deformable manipulation due to two challenges: (1) the state space is high-dimensional with physics-induced constraints, making valid configurations impossible to reach via low-dimensional pose perturbations; and (2) trajectory transfer is non-equivariant, as material points no longer move rigidly together under [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":[{"type":"text","value":"Zili Lin","user_id":0},{"type":"text","value":"Wenyao Zhang","user_id":0},{"type":"text","value":"Yuyang Zhang","user_id":0},{"type":"text","value":"Zekun Qi","user_id":0},{"type":"text","value":"Junyan Lin","user_id":0},{"type":"text","value":"Hanxin Zhu","user_id":0},{"type":"user_nicename","value":"Jiaolong Yang","user_id":"36125"},{"type":"text","value":"Zhibo Chen","user_id":0},{"type":"text","value":"Yao Mu","user_id":0},{"type":"text","value":"Xiaokang Yang","user_id":0},{"type":"user_nicename","value":"Xin Jin","user_id":"41958"},{"type":"user_nicename","value":"Wenjun Zeng","user_id":"34830"}],"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"arXiv","msr_isbn":"","msr_issue":"","msr_journal":"","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"","msr_copyright":"","msr_conference_name":"","msr_doi":"","msr_arxiv_id":"2606.25939","msr_mag_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_release_tracker_id":"","msr_highlight_type":"","msr_date_display_format":"","msr_main_download_label":"","msr_external_link_label":"","msr_doi_label":"","msr_published_date":"2026-06-24","msr_startdate":"","msr_presentation_date":"","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_year":2026,"msr_month":6,"msr_day":24,"msr_microsoftintellectualproperty":false,"msr_pub_id":"dbdc8d8a08da6a347493d938540354344d8996f0","msr_publication_uploader":[{"type":"url","title":"https:\/\/arxiv.org\/abs\/2606.25939","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_original_fields_of_study":[],"msr_s2_paper_id":"dbdc8d8a08da6a347493d938540354344d8996f0","msr_s2_pdf_url":"","msr_citation_count_updated":"","msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":65,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[{"provider":"s2","id":"dbdc8d8a08da6a347493d938540354344d8996f0"},{"provider":"arxiv","id":"2606.25939"},{"provider":"corpusid","id":"289628623"}],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13556,13552],"msr-publication-type":[270373],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[256039,246691,249835],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1177606","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-hardware-devices","msr-locale-en_us","msr-field-of-study-artifcial-intelligence","msr-field-of-study-computer-science","msr-field-of-study-robotics"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2026-06-24","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"arXiv","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":0,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","title":"https:\/\/arxiv.org\/abs\/2606.25939","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"dbdc8d8a08da6a347493d938540354344d8996f0","msr_influential_citations":0,"msr_reference_count":65,"msr_arxiv_id":"2606.25939","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Zili Lin","user_id":0,"rest_url":false},{"type":"text","value":"Wenyao Zhang","user_id":0,"rest_url":false},{"type":"text","value":"Yuyang Zhang","user_id":0,"rest_url":false},{"type":"text","value":"Zekun Qi","user_id":0,"rest_url":false},{"type":"text","value":"Junyan Lin","user_id":0,"rest_url":false},{"type":"text","value":"Hanxin Zhu","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Jiaolong Yang","user_id":36125,"rest_url":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jiaolong Yang"},{"type":"text","value":"Zhibo Chen","user_id":0,"rest_url":false},{"type":"text","value":"Yao Mu","user_id":0,"rest_url":false},{"type":"text","value":"Xiaokang Yang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Xin Jin","user_id":41958,"rest_url":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Xin Jin"},{"type":"user_nicename","value":"Wenjun Zeng","user_id":34830,"rest_url":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Wenjun Zeng"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"misc","related_content":[],"_links":{"self":[{"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1177606","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1177606\/revisions"}],"predecessor-version":[{"id":1178541,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1177606\/revisions\/1178541"}],"wp:attachment":[{"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1177606"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1177606"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1177606"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1177606"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1177606"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1177606"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1177606"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1177606"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1177606"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1177606"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1177606"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1177606"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1177606"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/find.codeghost.online\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1177606"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}