LGSep 25, 2025

A circuit for predicting hierarchical structure in-context in Large Language Models

arXiv:2509.21534v11 citationsh-index: 10
Originality Highly original
AI Analysis

This addresses a key limitation in understanding LLM in-context learning for complex patterns like those in natural language, offering a mechanistic account.

The study tackled the problem of whether LLMs can learn hierarchical patterns in-context, finding that adaptive induction heads enable accurate prediction by learning what to attend to, with evidence from synthetic tasks and natural language analogues.

Large Language Models (LLMs) excel at in-context learning, the ability to use information provided as context to improve prediction of future tokens. Induction heads have been argued to play a crucial role for in-context learning in Transformer Language Models. These attention heads make a token attend to successors of past occurrences of the same token in the input. This basic mechanism supports LLMs' ability to copy and predict repeating patterns. However, it is unclear if this same mechanism can support in-context learning of more complex repetitive patterns with hierarchical structure. Natural language is teeming with such cases: The article "the" in English usually prefaces multiple nouns in a text. When predicting which token succeeds a particular instance of "the", we need to integrate further contextual cues from the text to predict the correct noun. If induction heads naively attend to all past instances of successor tokens of "the" in a context-independent manner, they cannot support this level of contextual information integration. In this study, we design a synthetic in-context learning task, where tokens are repeated with hierarchical dependencies. Here, attending uniformly to all successor tokens is not sufficient to accurately predict future tokens. Evaluating a range of LLMs on these token sequences and natural language analogues, we find adaptive induction heads that support prediction by learning what to attend to in-context. Next, we investigate how induction heads themselves learn in-context. We find evidence that learning is supported by attention heads that uncover a set of latent contexts, determining the different token transition relationships. Overall, we not only show that LLMs have induction heads that learn, but offer a complete mechanistic account of how LLMs learn to predict higher-order repetitive patterns in-context.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes