arXiv cs.CL· Yuchen Li, Xintong Li, Zongqi Fan, Ken-tye Yong, Nguyen H. Tran·· 2 天前AI 评分24
大语言模型的“自我”在哪里？——从自我索引到自我指涉
Where Is the "Self" of a Large Language Model? -- From Self-Index to Self-Reference
AI 导读
研究在 Qwen3.8-27B 上区分了自我索引与自我指涉两个功能层级：将同一文本分别置于 Assistant、User、System 消息中，模型在全部 64 层、所有测试文本上都能沿共享方向区分自身消息与外部来源。
正文 · 原文
Abstract:Does a large language model have a self, and if so, of what kind? We separate two functional levels familiar from biology: a self-index, which distinguishes one's own material from external material and shapes processing, as corollary discharge does in nearly all animals, and self-reference, the higher ability to recognize what is one's own, as in mirror self-recognition. We probe both in Qwen3.8-27B by placing the same text in an Assistant, a User, or a System message rendered by the model's own chat template, and report three levels of evidence. Persistence: as predicted in advance, the representation of the text separates the model's own message from both external sources at all 64 layers and for every test text, along a shared direction, even when the answer does not depend on who supplied the text. Coupling: asked who wrote the preceding message, the model claims its own messages (0.77) and assigns User and System messages alike to the user, and how strongly it claims each message is predicted by the size of its self-index (Spearman correlation up to 0.68). Intervention: exchanging the internal states that follow a message, rather than those of its text, reverses the self-attribution between layers 15 and 31; removing the self-index direction from them at layer 39 lowers the claim more than a random direction; and in arithmetic the model adopts a wrong step more often when it wrote the step itself (0.92 versus 0.75), an effect carried by the step's representation at layers 31 to 39. Reasoning first in thinking mode raises the probability of the correct answer (0.96 instead of 0.77), but recognition is present without it. We propose that self-reference is a higher-level ability built on a self-index, a content-independent displacement superimposed on the whole representation of the model's own messages.
| Comments: | 20 pages, 8 figures |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.16717 [cs.CL] |
| (or arXiv:2610.16717v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16717 arXiv-issued DOI via DataCite |
Submission history
From: Yuchen Li [view email]
[v1]
Mon, 12 Oct 2026 12:51:04 UTC (2,313 KB)
[v2]
Thu, 15 Oct 2026 06:38:57 UTC (277 KB)
来源：arXiv cs.CL · arxiv.org