arXiv cs.CL· Munachiso Samuel Nwadike, Zangir Iklassov, Ali Mekky, Zayd M. Kawakibi Zuhri, Kentaro Inui·· 10 小时前AI 评分26
Einstein World Models：让 LLM 在推理链中做视觉思想实验
Einstein World Models
AI 导读
研究者提出 Einstein World Models（EWM），一种让 LLM 在推理轨迹中嵌入视觉-时序 rollout 的推理系统蓝图。LLM 调用 world-module 生成场景的短 rollout，返回结果被当作可检查的假设而非最终答案，从而把工具调用能力扩展到视觉思想实验领域。论文共 12 页，含 2 张图和 1 个算法。
正文 · 原文
Abstract:Does intelligence require the ability to reason about phenomena beyond direct experience? It is natural to suspect that some complex thought cannot be captured through language alone. However, of particular concern to this work, is whether visualising counterfactual events can complement language as a mechanism for complex thought. We ask whether LLMs can be trained to utilise such visualisation mechanisms, in a way that benefits their reasoning abilities. Motivated by this question, we propose Einstein World Models. EWMs are a blueprint for LLM-based reasoning systems that place visual-temporal rollouts inside the reasoning trace, allowing them to reason in ways that text alone may not support well. In an EWM, the LLM calls a world-module (not to be confused with a world model), to produce short rollouts of scenes under consideration. The returned rollout is treated not as the answer, but as an inspectable hypothesis that can support later reasoning. Einstein World Models extend the capability of LLMs for tool calling (such as web search or code execution), into the domain of visual thought experiments.
| Comments: | 12 pages (9 without references), 2 figures, 1 algorithm |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2606.26969 [cs.AI] |
| (or arXiv:2606.26969v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2606.26969 arXiv-issued DOI via DataCite |
Submission history
From: Munachiso Nwadike [view email]
[v1]
Thu, 25 Jun 2026 12:42:04 UTC (1,478 KB)
[v2]
Tue, 10 Nov 2026 00:19:48 UTC (1,479 KB)
来源：arXiv cs.CL · arxiv.org