{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T10:44:22Z","timestamp":1766400262965,"version":"3.48.0"},"reference-count":30,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iros60139.2025.11246704","type":"proceedings-article","created":{"date-parts":[[2025,11,27]],"date-time":"2025-11-27T18:54:45Z","timestamp":1764269685000},"page":"15873-15879","source":"Crossref","is-referenced-by-count":1,"title":["Disentangled Object-Centric Image Representation for Robotic Manipulation"],"prefix":"10.1109","author":[{"given":"David","family":"Emukpere","sequence":"first","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Romain","family":"Deffayet","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bingbing","family":"Wu","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Romain","family":"Br\u00e9gier","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Niemaz","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jean-Luc","family":"Meunier","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Denys","family":"Proux","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jean-Michel","family":"Renders","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Seungsu","family":"Kim","sequence":"additional","affiliation":[{"name":"NAVER LABS Europe,France,38240"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.026"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1177\/02783649241273668"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.016"},{"article-title":"Mobile aloha: Learning bimanual mobile manipulation with low-cost whole-body teleoperation","volume-title":"Conference on Robot Learning (CoRL)","author":"Fu","key":"ref4"},{"key":"ref5","first-page":"26 991","article-title":"Behavior generation with latent actions","volume-title":"Proceedings of the 41st International Conference on Machine Learning","volume":"235","author":"Lee"},{"article-title":"Masked visual pre-training for motor control","year":"2022","author":"Xiao","key":"ref6"},{"article-title":"Vrl3: A data-driven framework for visual deep reinforcement learning","volume-title":"Conference on Neural Information Processing Systems","author":"Wang","key":"ref7"},{"key":"ref8","first-page":"12 511","article-title":"On pre-training for visuo-motor control: Revisiting a learning-from-scratch baseline","volume-title":"Proceedings of the 40th International Conference on Machine Learning","volume":"202","author":"Hansen"},{"key":"ref9","first-page":"5221","article-title":"Generalization and robustness implications in object-centric learning","volume-title":"Proceedings of the 39th International Conference on Machine Learning","volume":"162","author":"Dittadi"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10610695"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.2974707"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2013.50"},{"article-title":"Towards a definition of disentangled representations","year":"2018","author":"Higgins","key":"ref13"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2024.3420937"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/lra.2022.3141148"},{"article-title":"Know thyself: Transferable visual control policies through robot-awareness","volume-title":"International Conference on Learning Representations","author":"Hu","key":"ref16"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2018.8461196"},{"key":"ref18","first-page":"11 525","article-title":"Object-centric learning with slot attention","volume":"33","author":"Locatello","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref19","first-page":"148","article-title":"Sornet: Spatial object-centric representations for sequential manipulation","volume-title":"Conference on Robot Learning","author":"Yuan"},{"article-title":"M2t2: Multi-task masked transformer for object-centric pick and place","volume-title":"Conference on Robot Learning","author":"Yuan","key":"ref20"},{"key":"ref21","first-page":"1199","article-title":"Viola: Imitation learning for vision-based manipulation with object proposal priors","volume-title":"Conference on Robot Learning","author":"Zhu"},{"volume-title":"Reinforcement Learning: An Introduction","year":"2018","author":"Sutton","key":"ref22"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1002\/SERIES1345"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.12794\/metadc1505267"},{"key":"ref25","article-title":"Maniskill 3.0.0.b5 documentation"},{"article-title":"Dinov2: Learning robust visual features without supervision","year":"2023","author":"Oquab","key":"ref26"},{"article-title":"Sam 2: Segment anything in images and videos","year":"2024","author":"Ravi","key":"ref27"},{"article-title":"R3m: A universal visual representation for robot manipulation","volume-title":"Conference on Robot Learning","author":"Nair","key":"ref28"},{"article-title":"Vip: Towards universal visual reward and representation via value-implicit pre-training","year":"2022","author":"Ma","key":"ref29"},{"article-title":"Liv: Language-image representations and rewards for robotic control","volume-title":"International Conference on Machine Learning","author":"Ma","key":"ref30"}],"event":{"name":"2025 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","start":{"date-parts":[[2025,10,19]]},"location":"Hangzhou, China","end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11245651\/11245652\/11246704.pdf?arnumber=11246704","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,18]],"date-time":"2025-12-18T12:37:02Z","timestamp":1766061422000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11246704\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":30,"URL":"https:\/\/doi.org\/10.1109\/iros60139.2025.11246704","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}