Abstract:World models have been recently proposed as sandbox environments in which AI agents can be trained and evaluated before deployment. While realistic world models often have high computational demands, this can often be alleviated by exploiting the fact that real-world scenarios tend to involve subcomponents that interact in a modular manner. In this paper, we explore this idea by developing a framework for decomposing complex world models represented by transducers, a class of models generalising POMDPs. Whereas the composition of transducers is well understood, our results clarify how to invert this process by deriving sub-transducers operating on distinct input-output subspaces, enabling parallelizable and interpretable alternatives to monolithic world modelling that can support distributed inference. Overall, these results lay groundwork for bridging the computational efficiency required for real-world inference and the structural transparency demanded by AI safety.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2512.02193 [cs.AI] |
| (or arXiv:2512.02193v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2512.02193 arXiv-issued DOI via DataCite |
From: Alexander Boyd [view email]
[v1]
Mon, 1 Dec 2025 20:37:43 UTC (5,435 KB)
[v2]
Wed, 20 May 2026 19:51:01 UTC (6,539 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。