Paper Detail
Alexander Pondaven, Ziyi Wu, Igor Gilitschenski, Philip Torr, Sergey Tulyakov, Fabio Pizzati, Aliaksandr Siarohin
Recent advances in video diffusion have enabled the development of "world models" capable of simulating interactive environments. However, these models are largely restricted to single-agent settings, failing to control multiple agents simultaneously in a scene. In this work, we tackle a fundamental issue of action binding in existing video diffusion models, which struggle to associate specific actions with their corresponding subjects. For this purpose, we propose ActionParty, an action controllable multi-subject world model for generative video games. It introduces subject state tokens, i.e. latent variables that persistently capture the state of each subject in the scene. By jointly modeling state tokens and video latents with a spatial biasing mechanism, we disentangle global video frame rendering from individual action-controlled subject updates. We evaluate ActionParty on the Melting Pot benchmark, demonstrating the first video world model capable of controlling up to seven players simultaneously across 46 diverse environments. Our results show significant improvements in action-following accuracy and identity consistency, while enabling robust autoregressive tracking of subjects through complex interactions.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{pondaven2026actionparty,
title = {ActionParty: Multi-Subject Action Binding in Generative Video Games},
author = {Alexander Pondaven and Ziyi Wu and Igor Gilitschenski and Philip Torr and Sergey Tulyakov and Fabio Pizzati and Aliaksandr Siarohin},
year = {2026},
abstract = {Recent advances in video diffusion have enabled the development of "world models" capable of simulating interactive environments. However, these models are largely restricted to single-agent settings, failing to control multiple agents simultaneously in a scene. In this work, we tackle a fundamental issue of action binding in existing video diffusion models, which struggle to associate specific actions with their corresponding subjects. For this purpose, we propose ActionParty, an action control},
url = {https://arxiv.org/abs/2604.02330},
keywords = {cs.CV, cs.AI, cs.LG, video diffusion, world models, action binding, multi-subject world model, subject state tokens, video latents, spatial biasing mechanism, autoregressive tracking, Melting Pot benchmark, code available, huggingface daily, Computer science, Rendering (computer graphics), Generative grammar, Action recognition, Artificial intelligence},
eprint = {2604.02330},
archiveprefix = {arXiv},
}
{}