AI 中文总结
研究语言模型中信念与现实分离的位置，核心方法是通过通用值槽和查询位置路由器，表明其基于两个可分离机制处于两个位置，结果在三种架构及多个模型家族成立，深入研究了信念 - 现实轴且该格式在多语境共享。
AI 中文摘要
有能力的语言模型能够区分角色所相信的与实际情况：当被告知“安娜认为杯子是蓝色的；实际上它是红色的”时，它们会回答安娜认为是蓝色，而实际是红色。这种分离在计算的哪个位置呢？我们表明它基于两个可分离的机制处于两个位置。一个通用值槽绑定属性值，查询位置的路由器选择查询读出的是角色的信念还是现实的框架。两条路由填充该槽：断言信念直接绑定文本提供的值，派生信念通过基于可见性的回溯到达。在一条路由上训练的子空间会引导另一条，只有派生路由依赖于描述的可见性。槽本身不携带信念 - 现实标签，分离存在于一对分离的路由子空间中，它们在框架之间翻转查询而不注入源的值。这些结果在三种架构上成立，在针对心理理论基准捷径进行去混淆的刺激上也成立；这种行为在五个模型家族的3B到7B模型之间出现。本文深入研究了单一的信念 - 现实轴；一篇配套论文表明相同的槽和路由器格式在句子可以开启的其他非实际语境（反事实、虚构、时间）中也共享。
英文摘要
Capable language models hold what a character believes apart from what is true: told "Anna believes the cup is blue; in reality it is red," they answer blue about Anna and red about the world. Where in the computation does that separation live? We show it rests on two separable mechanisms at two positions. A generic value slot binds the attributed value. A router at the query position selects which frame, the character's belief or reality, a query reads out. Two routes fill the slot: an asserted belief, whose value the text supplies, binds in directly; a derived belief, whose value must be inferred from what the character could see, arrives by a visibility-gated lookback. A subspace trained on either route steers the other, and only the derived route depends on described visibility. The slot itself carries no belief-reality tag: intervening on it moves a reality readout as strongly as a belief one. The separation lives instead in a dissociated pair of routing subspaces, which flip a query between frames without injecting the donor's value. These results hold across three architectures, on stimuli de-confounded against theory-of-mind-benchmark shortcuts; the behavior itself emerges between 3B and 7B across five model families. This paper develops the single belief-reality axis in depth; a companion paper shows the same slot-and-router format is shared across the other non-actual contexts a sentence can open (counterfactual, fictional, temporal).
Comments21 pages, 6 figures, 6 tables. v2: cite the released Mental Spaces Corpus (dataset DOI); switch to ACL bibliography style; minor copy-editing. No change to results
