Newest papers

OverSeer: Multi-Agent World Models Beyond What Agents See

Rui Li, Kam Woh Ng, Minghao Chen, Junlin Han, Jianyuan Wang, Zhaochong An, Juan-Manuel Pérez-Rúa, Andrea Vedaldi, Shikun Liu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28554 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Terlepas dari kesuksesan model dunia video yang efisien, memperluas mereka ke banyak agen sulit karena membutuhkan menghasilkan beberapa aliran video khusus agen yang konsisten dalam waktu dan antara agen. Untuk mengatasi kesenjangan ini, kami memperkenalkan OverSeer, model dunia multi-agent yang meningkatkan pandangan agen egosentris dengan pandangan kamera auxiliary. Kami menyarankan bahwa, dengan memperbaiki pandangan auxiliary di ruang, mereka bertindak sebagai proxy untuk representasi yang berpusat di dunia: Tidak seperti pandangan agen, yang dapat berubah sepenuhnya karena perubahan posisi agen sederhana, perubahan dalam pandangan auxiliary memetakan perubahan langsung ke perubahan lokal di negara dunia. Kami menunjukkan bahwa pandangan tetap ini secara substansial membantu model untuk mempertahankan konsistensi. Kami juga menunjukkan bahwa, karena sifat redundant dan semi-static mereka, pandangan auxiliary dapat dikompres secara efisien dengan delta-VAE yang hanya meng

Machine translation (Meta M2M100); the English original follows.

Abstract

Despite the success of video world models, extending them to multiple agents is difficult because it requires generating several agent-specific video streams that are consistent in time and between agents. To address this gap, we introduce OverSeer, a multi-agent world model that augments egocentric agent views with auxiliary camera views. We suggest that, by fixing the auxiliary views in space, they act as a proxy for a world-centric representation: Unlike the agents' views, which can change entirely due to a simple change in agent position, changes in the auxiliary views directly map to localized changes in the world state. We show that these fixed views substantially help the model to maintain consistency. We also show that, due to their redundant and semi-static nature, the auxiliary views can be efficiently compressed with a delta-VAE that encodes only the changes relative to a static reference image, reducing the number of tokens required to process them. Experiments in Minecraft and Doom demonstrate improved world and agent consistency, with qualitative results on real-world multi-camera captures indicating that the approach is not tied to game engines. Our compression analysis further shows that delta-VAE substantially reduces the computational overhead of auxiliary-view generation. We conclude that these fixed views provide a simple and efficient way to improve multi-agent world models.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Simple (philosophy), Process (computing), Overhead (engineering), Real world data, Proxy (statistics)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange