arXiv cs.AI· Sevan Brodjian, Michael Hobley, Pietro Perona·· 7 天前AI 评分29
生成模型何时能充当模拟器：可编辑性研究
When Generative Models Act as Simulators
AI 导读
研究探讨生成模型能否通过"可编辑性"充当模拟器，即改变模型潜在表示能否产生与编辑模拟器状态相同的下一步输出。作者在离散棋盘游戏 Othello 和自建连续世界 Rayworld 中测试了探针解码与逆映射两种方式，发现哪种映射成功强烈依赖环境：Othello 中可编辑性随游戏简化平滑下降，微小规则改动即产生巨大影响；Rayworld 中探针无法沿模拟器连续状态引导，但逆映射仍能设定模型状态。
正文 · 原文
Abstract:A simulator summarizes an environment in a state that users can read and change. Generative models learn representations of the world without explicit specification, but offer no built-in way to edit that world's state. We investigate the conditions under which a generative model can act as a simulator through $\textit{editability}$: whether changing a model's latent representation produces the next-step outputs we get by editing a simulator's state. Plainly stated, our goal is to find a mapping between a generative model's latent state and the simulator state, where the model is trained on observations alone and the mapping is learned afterward. We test two kinds of mappings, both learned from observations paired with simulator states. In one, we train probes to decode the simulator state from the model, then steer along them. In the other, we learn the mapping directly with an inverse map, which can succeed even where steering fails. We study both in two contrasting environments: the discrete board game Othello, and Rayworld, our own continuous world of moving discs seen through partial renderings, closer to physical scenes and video. We find that which mapping succeeds depends strongly on the environment, even though the state is decodable in every variant. In Othello, editability falls smoothly as the game is simplified, and even small rule changes have large consequences. In Rayworld, probes fail to steer along the simulator's continuous state, yet an inverse map sets the model from that state anyway. A model can therefore be driven by a simulator it may have never learned.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.12847 [cs.LG] |
| (or arXiv:2610.12847v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.12847 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sevan Brodjian [view email]
[v1]
Fri, 9 Oct 2026 01:12:21 UTC (2,047 KB)
来源：arXiv cs.AI · arxiv.org