MEGA Hub

Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction

Authors

Do you know Dahyun Chung?You can claim authorship or link another user.Do you know Siyoon Jin?You can claim authorship or link another user.Do you know Hyunwook Choi?You can claim authorship or link another user.Do you know Honggyu An?You can claim authorship or link another user.Do you know Junyoung Seo?You can claim authorship or link another user.Do you know Hyunsung Kim?You can claim authorship or link another user.Do you know Seung Wook Kim?You can claim authorship or link another user.Do you know Seungryong Kim?You can claim authorship or link another user.

Abstract

Egocentric world models predict first-person observations conditioned on an agent's actions, but most focus on a single agent. Real embodied settings often involve multiple agents that act and interact within a shared environment. Existing multi-agent world models rely on coarse actions like locomotion, camera control, or discrete commands, leaving fine-grained embodied interactions underexplored. We formulate multi-agent egocentric world modeling as synchronized ego-stream generation for multiple agents interacting through fine-grained actions in a shared world. This requires cross-view action consistency, shared-environment consistency, and consistent propagation of interaction-induced state updates. We propose Multi-agent Egocentric World Model (ME-World), which jointly denoises multiple ego streams in a shared token sequence, conditions each stream on all agents' target-view poses, and grounds generation with shared environment memory. We train and evaluate on real and synthetic multi-agent data and introduce shared-world consistency metrics for environment, update, and identity consistency. Experiments show ME-World improves shared-world consistency, action control, identity preservation, and video quality over existing methods.

Community

00