{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T04:22:42Z","timestamp":1750220562592,"version":"3.41.0"},"publisher-location":"New York, NY, USA","reference-count":84,"publisher":"ACM","license":[{"start":{"date-parts":[[2021,9,6]],"date-time":"2021-09-06T00:00:00Z","timestamp":1630886400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2021,9,6]]},"DOI":"10.1145\/3475946.3480950","type":"proceedings-article","created":{"date-parts":[[2021,9,9]],"date-time":"2021-09-09T22:05:42Z","timestamp":1631225142000},"page":"1-12","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["DASH: Modularized Human Manipulation Simulation with Vision and Language for Embodied AI"],"prefix":"10.1145","author":[{"given":"Yifeng","family":"Jiang","sequence":"first","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michelle","family":"Guo","sequence":"additional","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiangshan","family":"Li","sequence":"additional","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ioannis","family":"Exarchos","sequence":"additional","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiajun","family":"Wu","sequence":"additional","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"C. Karen","family":"Liu","sequence":"additional","affiliation":[{"name":"Stanford University, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2021,9,9]]},"reference":[{"key":"e_1_3_2_2_1_1","doi-asserted-by":"publisher","DOI":"10.1207\/s15327051hci1204_5"},{"volume-title":"Workshop on Virtual Reality Interaction and Physical Simulation, Jan Bender, Arjan Kuijper, Dieter\u00a0W","author":"Andrews Sheldon","key":"e_1_3_2_2_2_1"},{"key":"e_1_3_2_2_3_1","doi-asserted-by":"publisher","DOI":"10.1177\/0278364919887447"},{"key":"e_1_3_2_2_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/2933540.2933551"},{"volume-title":"Proceedings of the 2013 conference on empirical methods in natural language processing (EMNLP). 1533\u20131544","year":"2013","author":"Berant Jonathan","key":"e_1_3_2_2_5_1"},{"key":"e_1_3_2_2_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/70.897777"},{"key":"e_1_3_2_2_7_1","first-page":"1","article-title":"Introduction to inverse kinematics with jacobian transpose, pseudoinverse and damped least squares methods","volume":"17","author":"Buss R","year":"2004","journal-title":"IEEE Journal of Robotics and Automation"},{"volume-title":"Virtual Humanoids: Let them be autonomous without losing control. Technical Report","year":"2000","author":"Caicedo Angela","key":"e_1_3_2_2_8_1"},{"key":"e_1_3_2_2_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/332051.332075"},{"volume-title":"Computer Graphics Forum, Vol.\u00a039","author":"Chaudhuri Siddhartha","key":"e_1_3_2_2_10_1"},{"key":"e_1_3_2_2_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/344779.352172"},{"key":"e_1_3_2_2_12_1","unstructured":"Erwin Coumans and Yunfei Bai. 2016\u20132019. PyBullet a Python module for physics simulation for games robotics and machine learning. http:\/\/pybullet.org.  Erwin Coumans and Yunfei Bai. 2016\u20132019. PyBullet a Python module for physics simulation for games robotics and machine learning. http:\/\/pybullet.org."},{"key":"e_1_3_2_2_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3360905"},{"volume-title":"Assistive Gym: A Physics Simulation Framework for Assistive Robotics. arXiv preprint arXiv:1910.04700(2019).","year":"2019","author":"Erickson Zackory","key":"e_1_3_2_2_15_1"},{"key":"e_1_3_2_2_16_1","doi-asserted-by":"publisher","DOI":"10.1002\/cav.1560"},{"key":"e_1_3_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/311535.311538"},{"key":"e_1_3_2_2_18_1","unstructured":"Ross Girshick Ilija Radosavovic Georgia Gkioxari Piotr Doll\u00e1r and Kaiming He. 2018. Detectron. https:\/\/github.com\/facebookresearch\/detectron  Ross Girshick Ilija Radosavovic Georgia Gkioxari Piotr Doll\u00e1r and Kaiming He. 2018. Detectron. https:\/\/github.com\/facebookresearch\/detectron"},{"key":"e_1_3_2_2_19_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P17-1097"},{"key":"e_1_3_2_2_20_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-40415-3_33"},{"key":"e_1_3_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2018.8460699"},{"key":"e_1_3_2_2_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2010.5509683"},{"key":"e_1_3_2_2_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.322"},{"key":"e_1_3_2_2_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_3_2_2_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/1833349.1778770"},{"key":"e_1_3_2_2_26_1","unstructured":"Matthew Honnibal and Ines Montani. 2017. spaCy 2: Natural language understanding with Bloom embeddings convolutional neural networks and incremental parsing. (2017).  Matthew Honnibal and Ines Montani. 2017. spaCy 2: Natural language understanding with Bloom embeddings convolutional neural networks and incremental parsing. (2017)."},{"key":"e_1_3_2_2_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/1599470.1599476"},{"key":"e_1_3_2_2_28_1","unstructured":"Yuqian Jiang Shiqi Zhang Piyush Khandelwal and Peter Stone. 2018. Task Planning in Robotics: an Empirical Comparison of PDDL-based and ASP-based Systems. arXiv preprint arXiv:1804.08229(2018).  Yuqian Jiang Shiqi Zhang Piyush Khandelwal and Peter Stone. 2018. Task Planning in Robotics: an Empirical Comparison of PDDL-based and ASP-based Systems. arXiv preprint arXiv:1804.08229(2018)."},{"volume-title":"Eurographics 2011 - State of the Art Reports, N.\u00a0John and B.\u00a0Wyvill (Eds.)","author":"Jung Yvonne","key":"e_1_3_2_2_29_1"},{"key":"e_1_3_2_2_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/38.708560"},{"key":"e_1_3_2_2_31_1","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00101"},{"key":"e_1_3_2_2_32_1","doi-asserted-by":"crossref","unstructured":"Sujatha Krishnan-Barman Paul\u00a0AG Forbes and Antonia F de\u00a0C Hamilton. 2017. How can the study of action kinematics inform our understanding of human social interaction?Neuropsychologia 105(2017) 101\u2013110.  Sujatha Krishnan-Barman Paul\u00a0AG Forbes and Antonia F de\u00a0C Hamilton. 2017. How can the study of action kinematics inform our understanding of human social interaction?Neuropsychologia 105(2017) 101\u2013110.","DOI":"10.1016\/j.neuropsychologia.2017.01.018"},{"key":"e_1_3_2_2_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/1179352.1141969"},{"key":"e_1_3_2_2_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2000.844730"},{"volume-title":"The Hungarian method for the assignment problem. Naval research logistics quarterly 2, 1-2","year":"1955","author":"Kuhn W","key":"e_1_3_2_2_35_1"},{"key":"e_1_3_2_2_36_1","doi-asserted-by":"publisher","DOI":"10.5555\/2969442.2969523"},{"key":"e_1_3_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1007\/11821830_20"},{"key":"e_1_3_2_2_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3322972"},{"key":"e_1_3_2_2_39_1","doi-asserted-by":"publisher","DOI":"10.1162\/COLI_a_00127"},{"key":"e_1_3_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.106"},{"key":"e_1_3_2_2_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/1576246.1531365"},{"key":"e_1_3_2_2_42_1","doi-asserted-by":"publisher","DOI":"10.1177\/1729881419851402"},{"key":"e_1_3_2_2_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICAR.2011.6088576"},{"volume-title":"Interactive virtual humans in real-time virtual environments.IJVR 5, 2","year":"2006","author":"Magnenat-Thalmann Nadia","key":"e_1_3_2_2_44_1"},{"key":"e_1_3_2_2_45_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W16-3628"},{"key":"e_1_3_2_2_46_1","doi-asserted-by":"crossref","unstructured":"Oier Mees Alp Emek Johan Vertens and Wolfram Burgard. 2020. Learning Object Placements For Relational Instructions by Hallucinating Scene Representations. arXiv preprint arXiv:2001.08481(2020).  Oier Mees Alp Emek Johan Vertens and Wolfram Burgard. 2020. Learning Object Placements For Relational Instructions by Hallucinating Scene Representations. arXiv preprint arXiv:2001.08481(2020).","DOI":"10.1109\/ICRA40945.2020.9197472"},{"key":"e_1_3_2_2_47_1","unstructured":"Josh Merel Saran Tunyasuvunakool Arun Ahuja Yuval Tassa Leonard Hasenclever Vu Pham Tom Erez Greg Wayne and Nicolas Heess. 2019. Reusable neural skill embeddings for vision-guided whole body movement and object manipulation. arXiv preprint arXiv:1911.06636(2019).  Josh Merel Saran Tunyasuvunakool Arun Ahuja Yuval Tassa Leonard Hasenclever Vu Pham Tom Erez Greg Wayne and Nicolas Heess. 2019. Reusable neural skill embeddings for vision-guided whole body movement and object manipulation. arXiv preprint arXiv:1911.06636(2019)."},{"key":"e_1_3_2_2_48_1","doi-asserted-by":"publisher","DOI":"10.1145\/3386569.3392474"},{"key":"e_1_3_2_2_49_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793733"},{"key":"e_1_3_2_2_50_1","doi-asserted-by":"publisher","DOI":"10.5555\/2421731.2421752"},{"key":"e_1_3_2_2_51_1","unstructured":"Anusha Nagabandi Kurt Konoglie Sergey Levine and Vikash Kumar. 2019. Deep Dynamics Models for Learning Dexterous Manipulation. arXiv preprint arXiv:1909.11652(2019).  Anusha Nagabandi Kurt Konoglie Sergey Levine and Vikash Kumar. 2019. Deep Dynamics Models for Learning Dexterous Manipulation. arXiv preprint arXiv:1909.11652(2019)."},{"key":"e_1_3_2_2_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201305"},{"key":"e_1_3_2_2_53_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-21996-7_18"},{"volume-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV). 7588\u20137597","year":"2019","author":"Nguyen-Phuoc Thu","key":"e_1_3_2_2_54_1"},{"key":"e_1_3_2_2_55_1","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2000.844067"},{"key":"e_1_3_2_2_56_1","unstructured":"OpenAI Ilge Akkaya Marcin Andrychowicz Maciek Chociej Mateusz Litwin Bob McGrew Arthur Petron Alex Paino Matthias Plappert Glenn Powell Raphael Ribas 2019. Solving Rubik\u2019s Cube with a Robot Hand. arXiv preprint arXiv:1910.07113(2019).  OpenAI Ilge Akkaya Marcin Andrychowicz Maciek Chociej Mateusz Litwin Bob McGrew Arthur Petron Alex Paino Matthias Plappert Glenn Powell Raphael Ribas 2019. Solving Rubik\u2019s Cube with a Robot Hand. arXiv preprint arXiv:1910.07113(2019)."},{"key":"e_1_3_2_2_57_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8794441"},{"key":"e_1_3_2_2_58_1","doi-asserted-by":"publisher","DOI":"10.1145\/3272127.3275014"},{"key":"e_1_3_2_2_59_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-1402"},{"volume-title":"Computer Graphics Forum, Vol.\u00a021","author":"Peters Christopher","key":"e_1_3_2_2_60_1"},{"key":"e_1_3_2_2_61_1","doi-asserted-by":"publisher","DOI":"10.1145\/1073368.1073413"},{"key":"e_1_3_2_2_62_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00886"},{"key":"e_1_3_2_2_63_1","unstructured":"Tamer\u00a0F Rabie and Demetri Terzopoulos. 2000. Active perception in virtual humans. In Vision Interface Vol.\u00a02000.  Tamer\u00a0F Rabie and Demetri Terzopoulos. 2000. Active perception in virtual humans. In Vision Interface Vol.\u00a02000."},{"key":"e_1_3_2_2_64_1","doi-asserted-by":"crossref","unstructured":"Ilija Radosavovic Xiaolong Wang Lerrel Pinto and Jitendra Malik. 2020. State-Only Imitation Learning for Dexterous Manipulation. arXiv preprint arXiv:2004.04650(2020).  Ilija Radosavovic Xiaolong Wang Lerrel Pinto and Jitendra Malik. 2020. State-Only Imitation Learning for Dexterous Manipulation. arXiv preprint arXiv:2004.04650(2020).","DOI":"10.1109\/IROS51168.2021.9636557"},{"key":"e_1_3_2_2_65_1","doi-asserted-by":"crossref","unstructured":"Aravind Rajeswaran Vikash Kumar Abhishek Gupta Giulia Vezzani John Schulman Emanuel Todorov and Sergey Levine. 2017. Learning complex dexterous manipulation with deep reinforcement learning and demonstrations. arXiv preprint arXiv:1709.10087(2017).  Aravind Rajeswaran Vikash Kumar Abhishek Gupta Giulia Vezzani John Schulman Emanuel Todorov and Sergey Levine. 2017. Learning complex dexterous manipulation with deep reinforcement learning and demonstrations. arXiv preprint arXiv:1709.10087(2017).","DOI":"10.15607\/RSS.2018.XIV.049"},{"volume-title":"Animated agents for procedural training in virtual reality: Perception, cognition, and motor control. Applied artificial intelligence 13, 4-5","year":"1999","author":"Rickel Jeff","key":"e_1_3_2_2_66_1"},{"key":"e_1_3_2_2_67_1","doi-asserted-by":"publisher","DOI":"10.5555\/1170745.1171425"},{"key":"e_1_3_2_2_68_1","doi-asserted-by":"publisher","DOI":"10.5555\/3294771.3294871"},{"key":"e_1_3_2_2_69_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00943"},{"key":"e_1_3_2_2_70_1","unstructured":"John Schulman Filip Wolski Prafulla Dhariwal Alec Radford and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347(2017).  John Schulman Filip Wolski Prafulla Dhariwal Alec Radford and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347(2017)."},{"key":"e_1_3_2_2_71_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-25090-3_9"},{"key":"e_1_3_2_2_72_1","doi-asserted-by":"crossref","unstructured":"Mohit Shridhar and David Hsu. 2018. Interactive visual grounding of referring expressions for human-robot interaction. arXiv preprint arXiv:1806.03831(2018).  Mohit Shridhar and David Hsu. 2018. Interactive visual grounding of referring expressions for human-robot interaction. arXiv preprint arXiv:1806.03831(2018).","DOI":"10.15607\/RSS.2018.XIV.028"},{"key":"e_1_3_2_2_73_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01075"},{"key":"e_1_3_2_2_74_1","doi-asserted-by":"publisher","DOI":"10.5555\/3454287.3454388"},{"key":"e_1_3_2_2_75_1","doi-asserted-by":"publisher","DOI":"10.1162\/1064546053278973"},{"key":"e_1_3_2_2_76_1","doi-asserted-by":"crossref","unstructured":"Ayush Tewari Ohad Fried Justus Thies Vincent Sitzmann Stephen Lombardi Kalyan Sunkavalli Ricardo Martin-Brualla Tomas Simon Jason Saragih Matthias Nie\u00dfner Rohit Pandey Sean Fanello Gordon Wetzstein Jun-Yan Zhu Christian Theobalt Maneesh Agrawala Eli Shechtman Dan\u00a0B. Goldman and Michael Zollh\u00f6fer. 2020. State of the Art on Neural Rendering. Computer Graphics Forum(2020). https:\/\/doi.org\/10.1111\/cgf.14022  Ayush Tewari Ohad Fried Justus Thies Vincent Sitzmann Stephen Lombardi Kalyan Sunkavalli Ricardo Martin-Brualla Tomas Simon Jason Saragih Matthias Nie\u00dfner Rohit Pandey Sean Fanello Gordon Wetzstein Jun-Yan Zhu Christian Theobalt Maneesh Agrawala Eli Shechtman Dan\u00a0B. Goldman and Michael Zollh\u00f6fer. 2020. State of the Art on Neural Rendering. Computer Graphics Forum(2020). https:\/\/doi.org\/10.1111\/cgf.14022","DOI":"10.1111\/cgf.14022"},{"key":"e_1_3_2_2_77_1","doi-asserted-by":"publisher","DOI":"10.5555\/1402383.1402409"},{"key":"e_1_3_2_2_78_1","doi-asserted-by":"publisher","DOI":"10.1145\/2461912.2462000"},{"volume-title":"AllenAct: A Framework for Embodied AI Research. arXiv","year":"2020","author":"Weihs Luca","key":"e_1_3_2_2_79_1"},{"key":"e_1_3_2_2_80_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.744"},{"key":"e_1_3_2_2_81_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.2965078"},{"key":"e_1_3_2_2_82_1","doi-asserted-by":"publisher","DOI":"10.1145\/1186562.1015756"},{"key":"e_1_3_2_2_83_1","doi-asserted-by":"publisher","DOI":"10.1145\/2185520.2185537"},{"key":"e_1_3_2_2_84_1","doi-asserted-by":"publisher","DOI":"10.5555\/3326943.3327039"},{"key":"e_1_3_2_2_85_1","doi-asserted-by":"publisher","DOI":"10.1145\/2508363.2508412"}],"event":{"name":"SCA '21: The ACM SIGGRAPH \/ Eurographics Symposium on Computer Animation","sponsor":["SIGGRAPH ACM Special Interest Group on Computer Graphics and Interactive Techniques"],"location":"Virtual Event USA","acronym":"SCA '21"},"container-title":["The ACM SIGGRAPH \/ Eurographics Symposium on Computer Animation"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3475946.3480950","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3475946.3480950","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T21:28:33Z","timestamp":1750195713000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3475946.3480950"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,6]]},"references-count":84,"alternative-id":["10.1145\/3475946.3480950","10.1145\/3475946"],"URL":"https:\/\/doi.org\/10.1145\/3475946.3480950","relation":{},"subject":[],"published":{"date-parts":[[2021,9,6]]},"assertion":[{"value":"2021-09-09","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}