pith. sign in

arxiv: 2606.23296 · v1 · pith:M7RCDIGUnew · submitted 2026-06-22 · 💻 cs.RO

IOI: Decoupling Kinematics and Physics for Interactive World Models

Chengyu Bai , Peidong Jia , Tiecheng Guo , Yukai Wang , Rui Ma , Fangyuan Zhao , Chunkai Fan , Xiaobao Wei
show 6 more authors
Jintao Chen Hao Wang Ying Li Xiaozhu Ju Jian Tang Shanghang Zhang
This is my paper

Pith reviewed 2026-06-26 08:13 UTC · model grok-4.3

classification 💻 cs.RO
keywords interactive world modelskinematic priorsvideo generationembodied agentsrobot simulationpolicy evaluationzero-shot generalization
0
0 comments X

The pith

A hybrid interactive world model uses analytical kinematics to guide learned physics for accurate robot simulation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces IOI to build interactive world models for embodied agents by combining analytical forward kinematics with a learned video generator. It computes motion trajectories from actions and renders them as multi-view orthographic projections that are injected into the generator. This decoupling allows the model to maintain precise control alignment while modeling stochastic physical interactions. The approach leads to better simulation fidelity, zero-shot generalization to out-of-distribution tasks, and reliable policy evaluation that matches ground-truth simulators. Policies trained on data from this model perform comparably to those from real teleoperation in real-world settings.

Core claim

IOI integrates analytical kinematic priors with learned physical dynamics by computing forward kinematics from action sequences, rendering them into synchronized orthographic projections, and using a Multi-view Kinematic Aggregation and Injection module to provide geometry-consistent guidance to the video generator. This establishes synergy where the kinematic prior handles deterministic motion, freeing the generator to focus on physical interactions.

What carries the argument

The Multi-view Kinematic Aggregation and Injection module fuses orthographic projections of kinematic trajectories into the video generator to enforce geometry-consistent guidance.

If this is right

  • IOI achieves state-of-the-art simulation performance on the RoboTwin benchmark.
  • IOI enables robust zero-shot generalization to unseen out-of-distribution tasks.
  • IOI serves as a reliable policy evaluator with success rates aligning closely with ground-truth physics simulators.
  • Policies trained on IOI-synthesized data match those trained on teleoperation demonstrations when deployed on real-world platforms.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The separation of deterministic motion from stochastic dynamics could lower data needs for training world models by relying on analytical kinematics for trajectories.
  • Applying the same kinematic injection approach to non-rigid or multi-object scenes might extend reliable simulation beyond rigid-body robot tasks.
  • Using IOI as a policy evaluator could accelerate iteration in robotics by reducing dependence on full physics engines during early testing.

Load-bearing premise

The analytical kinematic model computes forward kinematics from action sequences with sufficient accuracy without needing extrinsic camera calibration.

What would settle it

If real-world tests show that policies trained on IOI-synthesized data achieve substantially lower success rates than those trained on teleoperation demonstrations, the claim of practical equivalence would be falsified.

read the original abstract

Developing generalist embodied agents requires interactive environments providing visually realistic feedback and accurate action-conditioned dynamics. Interactive world models address this by simulating such complex dynamics. However, purely data-driven methods struggle to ensure precise control alignment and physically plausible visual feedback due to a lack of explicit structural constraints. To address this, we propose IOI, a hybrid interactive world model integrating analytical kinematic priors with learned physical dynamics. Unlike data-driven approaches prone to spatiotemporal drift, IOI introduces explicit kinematic guidance, computing forward kinematics from action sequences for accurate motion trajectories. These trajectories are rendered into synchronized front, side, and top orthographic projections, eliminating the need for extrinsic camera calibration. A Multi-view Kinematic Aggregation and Injection module fuses these geometric cues and injects them into the video generator, providing geometry-consistent guidance. Conditioning video generation on these deterministic trajectories establishes a synergy between the analytical simulator and the world model. Decoupling deterministic motion into the kinematic prior frees the generator to model stochastic physical interactions. Experiments on the RoboTwin benchmark validate IOI across kinematic fidelity, out-of-distribution (OOD) generalization, and policy evaluation. IOI achieves state-of-the-art simulation performance and robust zero-shot generalization to unseen OOD tasks. Furthermore, IOI serves as a reliable policy evaluator, yielding success rates closely aligning with ground-truth physics simulators. On real-world platforms, policies trained on IOI-synthesized data match those trained on teleoperation demonstrations, solidifying its practical value for embodied policy learning.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The paper proposes IOI, a hybrid interactive world model that decouples kinematics from physics by computing analytical forward kinematics from action sequences, rendering them as synchronized front/side/top orthographic projections (eliminating extrinsic calibration), and injecting the fused geometric cues via a Multi-view Kinematic Aggregation and Injection module into a learned video generator. This is claimed to yield SOTA simulation fidelity on RoboTwin, robust zero-shot OOD generalization, success rates aligning with ground-truth simulators for policy evaluation, and real-world policies trained on IOI data matching those from teleoperation.

Significance. If the central decoupling holds with the claimed accuracy, the work would offer a concrete mechanism for injecting analytical structural priors into video-based world models, potentially improving long-horizon control alignment and physical plausibility over purely learned approaches. The reported real-world transfer and simulator alignment would strengthen its practical relevance for embodied policy learning.

major comments (3)
  1. [Abstract] Abstract and Experiments section: the central claim that explicit kinematic guidance enables SOTA performance and reliable policy evaluation rests on the accuracy of the analytical forward kinematics and multi-view fusion, yet no quantitative validation (reprojection error, trajectory drift over horizon length, or joint-limit violation rates) is reported; without these, it is impossible to confirm the prior is load-bearing rather than conditioning noise.
  2. [Abstract] Abstract: the claims of 'state-of-the-art simulation performance' and 'success rates closely aligning with ground-truth physics simulators' are presented without any numerical metrics, ablation tables, or error bars, which directly affects assessment of whether the decoupling produces the reported gains or whether results depend on benchmark-specific choices.
  3. [Kinematic guidance description] Kinematic prior description: the assertion that orthographic projections eliminate the need for extrinsic camera calibration is load-bearing for the geometry-consistent guidance claim, but no analysis of accumulation error in the forward kinematics computation or of the aggregation module's ability to enforce multi-view consistency in latent space is supplied.

Simulated Author's Rebuttal

3 responses · 0 unresolved

Thank you for the constructive feedback on our manuscript. We address each of the major comments point by point below, providing clarifications and indicating where revisions will be made to strengthen the presentation of the kinematic prior and experimental claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract and Experiments section: the central claim that explicit kinematic guidance enables SOTA performance and reliable policy evaluation rests on the accuracy of the analytical forward kinematics and multi-view fusion, yet no quantitative validation (reprojection error, trajectory drift over horizon length, or joint-limit violation rates) is reported; without these, it is impossible to confirm the prior is load-bearing rather than conditioning noise.

    Authors: We agree that explicit quantitative metrics on the kinematic prior would help demonstrate its contribution beyond conditioning. The manuscript validates kinematic fidelity on RoboTwin, but we acknowledge the absence of specific measures such as reprojection error and trajectory drift. In the revised version we will add these analyses, including accumulation over horizons and joint-limit checks, to confirm the prior is load-bearing. revision: yes

  2. Referee: [Abstract] Abstract: the claims of 'state-of-the-art simulation performance' and 'success rates closely aligning with ground-truth physics simulators' are presented without any numerical metrics, ablation tables, or error bars, which directly affects assessment of whether the decoupling produces the reported gains or whether results depend on benchmark-specific choices.

    Authors: The abstract is a concise summary; the full Experiments section contains the supporting numerical results, ablation tables, and error bars for SOTA comparisons and policy success rates. We will revise the abstract to include a few key quantitative values (with references to the tables) to make the claims more self-contained. revision: partial

  3. Referee: [Kinematic guidance description] Kinematic prior description: the assertion that orthographic projections eliminate the need for extrinsic camera calibration is load-bearing for the geometry-consistent guidance claim, but no analysis of accumulation error in the forward kinematics computation or of the aggregation module's ability to enforce multi-view consistency in latent space is supplied.

    Authors: Orthographic projections are generated directly from the analytical 3D forward-kinematics model, so no real-camera extrinsics are required; this is by construction. We agree that explicit analysis of accumulation error and latent-space multi-view consistency would strengthen the section. We will add this analysis, including quantitative checks on the aggregation module, in the revision. revision: yes

Circularity Check

0 steps flagged

No circularity: analytical kinematic prior is external and independent of learned generator

full rationale

The paper's core derivation uses an analytical forward-kinematics computation (explicitly described as deterministic and external) to produce orthographic projections that are then injected via a fusion module into a learned video generator. This separation is a modeling choice, not a self-referential definition or fitted parameter renamed as prediction. No equations or claims in the abstract reduce the output performance metrics to the input trajectories by construction, nor do any load-bearing steps rely on self-citations whose validity depends on the present work. The claimed decoupling therefore remains a substantive architectural hypothesis whose empirical support (SOTA numbers, OOD generalization, policy alignment) is independent of the derivation itself.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities. The kinematic prior is treated as given from analytical robotics rather than derived or fitted within the paper.

pith-pipeline@v0.9.1-grok · 5837 in / 1034 out tokens · 25158 ms · 2026-06-26T08:13:40.224063+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

41 extracted references · 1 canonical work pages

  1. [1]

    Cosmos world foundation model platform for physical ai.arXiv preprint arXiv:2501.03575, 2025

    Niket Agarwal, Arslan Ali, Maciej Bala, Yogesh Balaji, Erik Barker, Tiffany Cai, Prithvijit Chattopadhyay, Yongxin Chen, Yin Cui, Yifan Ding, et al. Cosmos world foundation model platform for physical ai.arXiv preprint arXiv:2501.03575, 2025

    Pith/arXiv arXiv 2025

  2. [2]

    Aloha 2: An enhanced low-cost hardware for bimanual teleoperation

    Jose Aldaco, Travis Armstrong, Robert Baruch, Jeff Bingham, Sean Chan, Kenneth Draper, Debidatta Dwibedi, Chelsea Finn, Pete Florence, Spencer Goodrich, et al. Aloha 2: An enhanced low-cost hardware for bimanual teleoperation. arXiv preprint arXiv:2405.02292, 2024

    arXiv 2024

  3. [3]

    The reality gap in robotics: Challenges, solutions, and best practices

    Elie Aljalbout, Jiaxu Xing, Angel Romero, Iretiayo Akinola, Caelan Reed Garrett, Eric Heiden, Abhishek Gupta, Tucker Hermans, Yashraj Narang, Dieter Fox, Davide Scaramuzza, and Fabio Ramos. The reality gap in robotics: Challenges, solutions, and best practices. Annual Review of Control, Robotics, and Autonomous Systems, 9, 2025

    2025

  4. [4]

    Self-supervised learning from images with a joint-embedding predictive architecture

    Mahmoud Assran, Quentin Duval, Ishan Misra, Piotr Bojanowski, Pascal Vincent, Michael Rabbat, Yann LeCun, and Nicolas Ballas. Self-supervised learning from images with a joint-embedding predictive architecture. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 15619–15629, 2023

    2023

  5. [5]

    V-jepa: Latent video prediction for visual representation learning

    Adrien Bardes, Quentin Garrido, Jean Ponce, Xinlei Chen, Michael Rabbat, Yann LeCun, Mido Assran, and Nicolas Ballas. V-jepa: Latent video prediction for visual representation learning. 2023

    2023

  6. [6]

    Motus: A unified latent action world model.arXiv preprint arXiv:2512.13030, 2025

    Hongzhe Bi, Hengkai Tan, Shenghao Xie, Zeyuan Wang, Shuhe Huang, Haitian Liu, Ruowen Zhao, Yao Feng, Chendong Xiang, Yinze Rong, et al. Motus: A unified latent action world model.arXiv preprint arXiv:2512.13030, 2025

    Pith/arXiv arXiv 2025

  7. [7]

    Kevin Black, Noah Brown, Danny Driess, Adnan Esmail, Michael Equi, Chelsea Finn, et al.π0: A vision-language- action flow model for general robot control.arXiv preprint arXiv:2410.24164, 2024

    Pith/arXiv arXiv 2024

  8. [8]

    Stable video diffusion: Scaling latent video diffusion models to large datasets.arXiv preprint arXiv:2311.15127, 2023

    Andreas Blattmann, Tim Dockhorn, Sumith Kulal, Daniel Mendelevitch, Maciej Kilian, Dominik Lorenz, Yam Levi, Zion English, Vikram Voleti, Adam Letts, et al. Stable video diffusion: Scaling latent video diffusion models to large datasets.arXiv preprint arXiv:2311.15127, 2023

    Pith/arXiv arXiv 2023

  9. [9]

    Rt-1: Robotics transformer for real-world control at scale

    Anthony Brohan, Noah Brown, Justice Carbajal, Yevgen Chebotar, Joseph Dabis, Chelsea Finn, Karol Hausman, et al. Rt-1: Robotics transformer for real-world control at scale. InRobotics: Science and Systems (RSS), 2023

    2023

  10. [10]

    Genie: Generative interactive environments

    Jake Bruce, Michael D Dennis, Ashley Edwards, Jack Parker-Holder, Yuge Shi, Edward Hughes, Matthew Lai, Aditi Mavalankar, Richie Steigerwald, Chris Apps, et al. Genie: Generative interactive environments. InForty-first International Conference on Machine Learning, 2024

    2024

  11. [11]

    Worldvla: Towards autoregressive action world model.arXiv preprint arXiv:2506.21539, 2025

    Jun Cen, Chaohui Yu, Hangjie Yuan, Yuming Jiang, Siteng Huang, Jiayan Guo, Xin Li, Yibing Song, Hao Luo, Fan Wang, et al. Worldvla: Towards autoregressive action world model.arXiv preprint arXiv:2506.21539, 2025

    Pith/arXiv arXiv 2025

  12. [12]

    Bridgev2w: Bridging video generation models to embodied world models via embodiment masks, 2026

    Yixiang Chen, Peiyan Li, Jiabing Yang, Keji He, Xiangnan Wu, Yuan Xu, Kai Wang, Jing Liu, Nianfeng Liu, Yan Huang, and Liang Wang. Bridgev2w: Bridging video generation models to embodied world models via embodiment masks, 2026

    2026

  13. [13]

    Ec-flow: Enabling versatile robotic manipulation from action-unlabeled videos via embodiment- centric flow

    Yixiang Chen et al. Ec-flow: Enabling versatile robotic manipulation from action-unlabeled videos via embodiment- centric flow. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025

    2025

  14. [14]

    Wow: Towards a world omniscient world model through embodied interaction, 2025

    Xiaowei Chi, Peidong Jia, Chun-Kai Fan, Xiaozhu Ju, Weishi Mi, Kevin Zhang, Zhiyuan Qin, Wanxin Tian, Kuangzhi Ge, Hao Li, Zezhong Qian, Anthony Chen, Qiang Zhou, Yueru Jia, Jiaming Liu, Yong Dai, Qingpo Wuwu, Chengyu Bai, Yu-Kai Wang, Ying Li, Lizhang Chen, Yong Bao, Zhiyuan Jiang, Jiacheng Zhu, Kai Tang, Ruichuan An, Yulin Luo, Qiuxuan Feng, Siyuan Zhou...

    2025

  15. [15]

    Motion prompting: Controlling video generation with motion trajectories

    Daniel Geng, Charles Herrmann, Junhwa Hur, Forrester Cole, Serena Zhang, Tobias Pfaff, Tatiana Lopez-Guevara, Yusuf Aytar, Michael Rubinstein, Chen Sun, Oliver Wang, Andrew Owens, and Deqing Sun. Motion prompting: Controlling video generation with motion trajectories. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVP...

    2025

  16. [16]

    Ctrl-world: A controllable generative world model for robot manipulation

    Yanjiang Guo, Lucy Xiaoyang Shi, Jianyu Chen, and Chelsea Finn. Ctrl-world: A controllable generative world model for robot manipulation. InThe FourteenthInternational Conference on Learning Representations (ICLR), 2026. 14

    2026

  17. [17]

    Ltx-video: Realtime video latent diffusion.arXiv preprint arXiv:2501.00103, 2024

    Yoav HaCohen, Nisan Chiprut, Benny Brazowski, Daniel Shalem, Dudu Moshe, Eitan Richardson, Eran Levin, Guy Shiran, Nir Zabari, Ori Gordon, et al. Ltx-video: Realtime video latent diffusion.arXiv preprint arXiv:2501.00103, 2024

    Pith/arXiv arXiv 2024

  18. [18]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. InNeurIPS, 2020

    2020

  19. [19]

    Vid2world: Crafting video diffusion models to interactive world models

    Siqiao Huang, Jialong Wu, Qixing Zhou, Shangchen Miao, and Mingsheng Long. Vid2world: Crafting video diffusion models to interactive world models. InInternational Conference on Learning Representations, 2026

    2026

  20. [20]

    Dreamgen: Unlocking generalization in robot learning through video world models

    Joel Jang, Seonghyeon Ye, Zongyu Lin, Jiannan Xiang, Johan Bjorck, Yu Fang, Fengyuan Hu, Spencer Huang, Kaushil Kundalia, Yen-Chen Lin, Loïc Magne, Ajay Mandlekar, Avnish Narayan, You Liang Tan, Guanzhi Wang, Jing Wang, Qi Wang, Yinzhen Xu, Xiaohui Zeng, Kaiyuan Zheng, Ruijie Zheng, Ming-Yu Liu, Luke Zettlemoyer, Dieter Fox, Jan Kautz, Scott Reed, Yuke Zh...

    2025

  21. [21]

    Vace: All-in-one video creation and editing

    Zeyinzi Jiang, Zhen Han, Chaojie Mao, Jingfeng Zhang, Yulin Pan, and Yu Liu. Vace: All-in-one video creation and editing. arXiv preprint arXiv:2503.07598, 2025

    Pith/arXiv arXiv 2025

  22. [22]

    OpenVLA: An open-source vision-language-action model

    Moo Jin Kim, Karl Pertsch, Siddharth Karamcheti, Ted Xiao, Ashwin Balakrishna, Suraj Nair, Rafael Rafailov, Ethan P Foster, Pannag R Sanketi, Quan Vuong, Thomas Kollar, Benjamin Burchfiel, Russ Tedrake, Dorsa Sadigh, Sergey Levine, Percy Liang, and Chelsea Finn. OpenVLA: An open-source vision-language-action model. In 8th Annual Conference on Robot Learning, 2024

    2024

  23. [23]

    Manipdreamer3d: Synthesizing plausible robotic manipulation video with occupancy-aware 3d trajectory

    Ying Li, Xiaobao Wei, Xiaowei Chi, Yuming Li, Zhongyu Zhao, Hao Wang, Ningning Ma, Ming Lu, and Sirui Han. Manipdreamer3d: Synthesizing plausible robotic manipulation video with occupancy-aware 3d trajectory. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 6644–6652, 2026

    2026

  24. [24]

    World model on million-length video and language with ringattention

    Hao Liu, Wilson Yan, Matei Zaharia, and Pieter Abbeel. World model on million-length video and language with ringattention. arXiv preprint arXiv:2402.08268, 2024

    Pith/arXiv arXiv 2024

  25. [25]

    Robocasa: Large-scale simulation of everyday tasks for generalist robots

    Soroush Nasiriany, Abhiram Maddukuri, Lance Zhang, Adeet Parikh, Aaron Lo, Abhishek Joshi, Ajay Mandlekar, and Yuke Zhu. Robocasa: Large-scale simulation of everyday tasks for generalist robots. InRobotics: Science and Systems (RSS), 2024

    2024

  26. [26]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 4195–4205, October 2023

    2023

  27. [27]

    Worldgym: World model as an environment for policy evaluation, 2025

    Julian Quevedo, Ansh Kumar Sharma, Yixiang Sun, Varad Suryavanshi, Percy Liang, and Sherry Yang. Worldgym: World model as an environment for policy evaluation, 2025

    2025

  28. [28]

    Avid: Adapting video diffusion models to world models,

    Marc Rigter, Tarun Gupta, Agrin Hilmkil, and Chao Ma. Avid: Adapting video diffusion models to world models,

  29. [29]

    URLhttps://arxiv.org/abs/2410.12822

    arXiv

  30. [30]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10684–10695, June 2022

    2022

  31. [31]

    Maniskill3: Gpu parallelized robotics simulation and rendering for generalizable embodied ai

    Stone Tao, Fanbo Xiang, Arth Shukla, Yuzhe Qin, Xander Hinrichsen, Xiaodi Yuan, Chen Bao, Xio Han, Jiayuan Wang, Taimin Mu, et al. Maniskill3: Gpu parallelized robotics simulation and rendering for generalizable embodied ai. In International Conference on Learning Representations (ICLR), 2025

    2025

  32. [32]

    Scalable policy evaluation with video world models, 2025

    Wei-Cheng Tseng, Jinwei Gu, Qinsheng Zhang, Hanzi Mao, Ming-Yu Liu, Florian Shkurti, and Lin Yen-Chen. Scalable policy evaluation with video world models, 2025. URLhttps://arxiv.org/abs/2511.11520

    arXiv 2025

  33. [33]

    Wan: Open and advanced large-scale video generative models.arXiv preprint arXiv:2503.20314, 2025

    Team Wan, Ang Wang, Baole Ai, Bin Wen, Chaojie Mao, Chen-Wei Xie, Di Chen, Feiwu Yu, Haiming Zhao, Jianx- iao Yang, et al. Wan: Open and advanced large-scale video generative models.arXiv preprint arXiv:2503.20314, 2025

    Pith/arXiv arXiv 2025

  34. [34]

    Kinema4d: Kinematic 4d world modeling for spatiotemporal embodied simulation, 2026

    Mutian Xu, Tianbao Zhang, Tianqi Liu, Zhaoxi Chen, Xiaoguang Han, and Ziwei Liu. Kinema4d: Kinematic 4d world modeling for spatiotemporal embodied simulation, 2026. URLhttps://arxiv.org/abs/2603.16669

    arXiv 2026

  35. [35]

    Learning interactive real-world simulators

    Sherry Yang, Yilun Du, Seyed Kamyar Seyed Ghasemipour, Jonathan Tompson, Leslie Pack Kaelbling, Dale Schuurmans, and Pieter Abbeel. Learning interactive real-world simulators. In The Twelfth International Conference on Learning Representations, 2024. URLhttps://openreview.net/forum?id=sFyTZEqmUY. 15

    2024

  36. [36]

    Cogvideox: Text-to-video diffusion models with an expert transformer

    Zhuoyi Yang, Jiayan Teng, Wendi Zheng, Ming Ding, Shiyu Huang, Jiazheng Xu, Yuanming Yang, Wenyi Hong, Xiaohan Zhang, Guanyu Feng, Da Yin, Yuxuan.Zhang, Weihan Wang, Yean Cheng, Bin Xu, Xiaotao Gu, Yuxiao Dong, and Jie Tang. Cogvideox: Text-to-video diffusion models with an expert transformer. InThe Thirteenth International Conference on Learning Represen...

    2025

  37. [37]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 3836–3847, 2023

    2023

  38. [38]

    Tesseract: Learning 4d embodied world models

    Haoyu Zhen, Qiao Sun, Hongxin Zhang, Junyan Li, Siyuan Zhou, Yilun Du, and Chuang Gan. Tesseract: Learning 4d embodied world models. InInternational Conference on Computer Vision (ICCV), 2025

    2025

  39. [39]

    Irasim: A fine-grained world model for robot manipulation

    Fangqi Zhu, Hongtao Wu, Song Guo, Yuxiao Liu, Chilam Cheang, and Tao Kong. Irasim: A fine-grained world model for robot manipulation. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025

    2025

  40. [40]

    Zhu, Pranav Kuppili, Ryan Punamiya, Patcharapong Aphiwetsa, Dhruv Patel, Simar Kareer, Sehoon Ha, and Danfei Xu

    Lawrence Y. Zhu, Pranav Kuppili, Ryan Punamiya, Patcharapong Aphiwetsa, Dhruv Patel, Simar Kareer, Sehoon Ha, and Danfei Xu. Emma: Scaling mobile manipulation via egocentric human data. IEEE Robotics and Automation Letters, 2025. doi: 10.1109/LRA.2025.11352854

    work page doi:10.1109/lra.2025.11352854 2025

  41. [41]

    RT-2: Vision-language- action models transfer web knowledge to robotic control

    Brianna Zitkovich, Tianhe Yu, Sichun Xu, Peng Xu, Ted Xiao, Fei Xia, Jialin Wu, et al. RT-2: Vision-language- action models transfer web knowledge to robotic control. In7th Annual Conference on Robot Learning, 2023. 16

    2023