[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84717-en":3,"doc-seo-84717-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84717,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","SoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects","SoftVTBench introduces a safety-aware visuo-tactile benchmark for physically constrained manipulation of deformable objects, addressing gaps in prior benchmarks that focus mainly on goal completion. Built in Isaac Sim with finite-element-simulated deformable objects, it supplies multi-view RGB, RGB tactile sensing with marker motion, proprioception, and language instructions. The benchmark reports both Goal Success and Safety Success, where safety enforces no-drop and bounded peak deformation using calibrated, privileged FEM signals. Experiments show safety-only evaluation reveals overstated performance, while tactile sensing improves Safety Success without reducing Goal Success.","arXiv :2607 .04234v 1 [ cs .RO] 5 Jul 2026  \nSoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects  \nBowen Jing 1 ,∗ , Mingxin Wang1 ,2 ,∗ , Ruiyang Hao3 , Chenchen Ge 1 ,4 , Hanwen Shen5 , Junjie He6 , Yang Cui7 , Yiming Hou 1 ,4 , Weitao Zhou2 ,8 ,‡, Jiawei Wang8 , Minglei Li8 , Dandan Zhang9 , Ding Zhao 10 , Houde Liu2 , Xiaofan Li 11 , Si Liu 12 , Ping Luo 13 , Haibao Yu 1 , 13 ,‡  \n1 Tuojing Intelligence, 2 Tsinghua University, 3 King’s College London, 4 Southeast University,  \n5 Stevens Institute of Technology, 6 The Hong Kong University of Science and Technology (Guangzhou),  \n7 University of Manchester, 8 Simple AI, 9 Imperial College London, 10 Carnegie Mellon University,  \n11 Zhejiang University, 12 Beihang University, 13 The University of Hong Kong  \n∗ equal contribution, ‡corresponding author  \nDeformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or drop while avoiding excessive deformation. However, existing manipulation benchmarks are predominantly success-oriented and rarely evaluate whether a policy remains physically safe throughout execution. We present SoftVTBench, a safety-aware visuo-tactile benchmark for physically constrained deformable object manipulation. Built in Isaac Sim with finite-element-simulated deformable objects, SoftVTBench provides multi-view RGB observations, RGB tactile sensing with marker motion, proprioception, and language instructions, and defines four matched task suites over object type (deformable vs. rigid) and variation axis (object vs. spatial) . It separately reports Goal Success and Safety Success; the latter additionally requires no drop and peak deformation below a calibrated object-specific threshold, measured from policy-hidden privileged Finite Element Method (FEM) states. We implement π0.5-based baselines under this protocol. Experiments show that success-only evaluation substantially overstates policy performance, as a large fraction of goal-completing rollouts still violate physical safety. Furthermore, incorporating tactile sensing improves Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation during execution, while maintaining comparable Goal Success. SoftVTBench provides a reproducible benchmark for studying visuo-tactile  \ndeformable manipulation under physical interaction constraints.  \nCode: [https://github.com/TuojingAI/SoftVTBench](https://github.com/TuojingAI/SoftVTBench)  \nWebsite: [https://softvtbench.github.io/](https://softvtbench.github.io/)  \n1 Introduction  \nRecent progress in large-scale robot learning and foundation-model-based policies has significantly improved the generality of robotic manipulation across tasks, objects, and environments Walke et al. (2023); Kim et al.(2024); Black et al. (2024); Bjorck et al. (2025); Jang et al. (2025); Ji et al. (2025) . However, most existing approaches are primarily evaluated in rigid-object settings Liu et al. (2023); Gu et al. (2023); Nasiriany et al.(2024); Mu et al. (2025), where performance is measured by goal achievement and the physical interaction process is largely abstracted away. Although deformable object manipulation has recently received increasing attention Zhao et al. (2025); Moletta et al. (2026); Moghani et al. (2026), existing evaluation protocols remain largely success-oriented, assessing performance primarily based on task completion Huang et al.(2021); Zhang et al. (2025c) . Unlike rigid-object manipulation, deformable object manipulation inherently involves contact-rich dynamics and material-dependent constraints Sun et al. (2025), in which successful execution requires not only accomplishing the task but also maintaining physically appropriate interactions, such as holding the object stably without slip or drop and avoiding excessive deformati","cbCaiplCPPArQ5Eo","https://ap.wps.com/l/cbCaiplCPPArQ5Eo","pdf",11425179,1,17,"English","en",105,"# Introduction\n## SoftVTBench Overview\n## Why Safety-Aware Evaluation Matters\n## Role of Visuo-Tactile Perception","[{\"question\":\"What problem does SoftVTBench target in deformable object manipulation benchmarks?\",\"answer\":\"It targets the limitation of success-only benchmarks that do not evaluate whether a policy remains physically safe throughout execution, including slip/drop and excessive deformation.\"},{\"question\":\"How does SoftVTBench define and measure Safety Success?\",\"answer\":\"Safety Success requires no drop and peak deformation below a calibrated, object-specific threshold, measured from privileged policy-hidden finite-element-method (FEM) states.\"},{\"question\":\"What impact does tactile sensing have on results in SoftVTBench?\",\"answer\":\"Incorporating tactile sensing increases Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation while keeping Goal Success comparable.\"}]",1784197818,43,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"softvtbench-a-safety-aware-visuo-tactile-benchmark-for-physically-constrained-robotic-manipulation-of-deformable-objects","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/softvtbench-a-safety-aware-visuo-tactile-benchmark-for-physically-constrained-robotic-manipulation-of-deformable-objects/84717/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does SoftVTBench target in deformable object manipulation benchmarks?","Question",{"text":75,"@type":76},"It targets the limitation of success-only benchmarks that do not evaluate whether a policy remains physically safe throughout execution, including slip/drop and excessive deformation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does SoftVTBench define and measure Safety Success?",{"text":80,"@type":76},"Safety Success requires no drop and peak deformation below a calibrated, object-specific threshold, measured from privileged policy-hidden finite-element-method (FEM) states.",{"name":82,"@type":73,"acceptedAnswer":83},"What impact does tactile sensing have on results in SoftVTBench?",{"text":84,"@type":76},"Incorporating tactile sensing increases Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation while keeping Goal Success comparable.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]