What Makes a Good Query? Measuring the Impact of Human-Confusing Linguistic Features on LLM Performance

TL;DR

This paper investigates how linguistic features of user queries influence hallucination rates in large language models, identifying specific features that increase or decrease hallucination risk to inform better query design.

Contribution

It introduces a 22-dimensional linguistic feature vector for queries and demonstrates how these features correlate with hallucination likelihood in large-scale LLM analysis.

Findings

Deep clause nesting increases hallucination risk

Clear intention grounding reduces hallucination

Domain-specific features have mixed effects

Abstract

Large Language Model (LLM) hallucinations are usually treated as defects of the model or its decoding strategy. Drawing on classical linguistics, we argue that a query's form can also shape a listener's (and model's) response. We operationalize this insight by constructing a 22-dimension query feature vector covering clause complexity, lexical rarity, and anaphora, negation, answerability, and intention grounding, all known to affect human comprehension. Using 369,837 real-world queries, we ask: Are there certain types of queries that make hallucination more likely? A large-scale analysis reveals a consistent "risk landscape": certain features such as deep clause nesting and underspecification align with higher hallucination propensity. In contrast, clear intention grounding and answerability align with lower hallucination rates. Others, including domain specificity, show mixed, dataset- and model-dependent effects. Thus, these findings establish an empirically observable query-feature representation correlated with hallucination risk, paving the way for guided query rewriting and future intervention studies.