Rhetorical Questions in LLM Representations: A Linear Probing Study

TL;DR

This study investigates how large language models internally represent rhetorical questions, revealing that multiple linear directions encode different rhetorical cues, with stable detection across datasets but diverse underlying phenomena.

Contribution

It provides the first detailed analysis of rhetorical question representations in LLMs, showing multiple encoding directions and the complexity of transferability.

Findings

Rhetorical signals emerge early in LLM representations.

Rhetorical questions are linearly separable from information-seeking questions.

Transferability varies, with different probes capturing distinct rhetorical phenomena.

Abstract

Rhetorical questions are asked not to seek information but to persuade or signal stance. How large language models internally represent them remains unclear. We analyze rhetorical questions in LLM representations using linear probes on two social-media datasets with different discourse contexts, and find that rhetorical signals emerge early and are most stably captured by last-token representations. Rhetorical questions are linearly separable from information-seeking questions within datasets, and remain detectable under cross-dataset transfer, reaching AUROC around 0.7-0.8. However, we demonstrate that transferability does not simply imply a shared representation. Probes trained on different datasets produce different rankings when applied to the same target corpus, with overlap among the top-ranked instances often below 0.2. Qualitative analysis shows that these divergences correspond to distinct rhetorical phenomena: some probes capture discourse-level rhetorical stance embedded in extended argumentation, while others emphasize localized, syntax-driven interrogative acts. Together, these findings suggest that rhetorical questions in LLM representations are encoded by multiple linear directions emphasizing different cues, rather than a single shared direction.