Latent Particle World Models: Self-Supervised Object-Centric Stochastic Dynamics Modeling
Abstract
We introduce Latent Particle World Model (LPWM), a self-supervised object-centric world model scaled to real-world multi-object datasets and applicable in decision-making. LPWM autonomously discovers keypoints, bounding boxes, and object masks directly from video data, enabling it to learn rich scene decompositions without supervision. Our architecture is trained end-to-end purely from videos and supports flexible conditioning on actions, language, and image goals. LPWM models stochastic particle dynamics via a novel latent action module and achieves state-of-the-art results on diverse real-world and synthetic datasets. Beyond stochastic video modeling, LPWM is readily applicable to decision-making, including goal-conditioned imitation learning, as we demonstrate in the paper. Code, data, pre-trained models and video rollouts are available: https://taldatech.github.io/lpwm-web
Cite
Text
Daniel et al. "Latent Particle World Models: Self-Supervised Object-Centric Stochastic Dynamics Modeling." International Conference on Learning Representations, 2026.Markdown
[Daniel et al. "Latent Particle World Models: Self-Supervised Object-Centric Stochastic Dynamics Modeling." International Conference on Learning Representations, 2026.](https://mlanthology.org/iclr/2026/daniel2026iclr-latent/)BibTeX
@inproceedings{daniel2026iclr-latent,
title = {{Latent Particle World Models: Self-Supervised Object-Centric Stochastic Dynamics Modeling}},
author = {Daniel, Tal and Qi, Carl and Haramati, Dan and Zadeh, Amir and Li, Chuan and Tamar, Aviv and Pathak, Deepak and Held, David},
booktitle = {International Conference on Learning Representations},
year = {2026},
url = {https://mlanthology.org/iclr/2026/daniel2026iclr-latent/}
}