Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs

TL;DR

This paper investigates how reasoning enhances parametric knowledge recall in large language models even for simple factual questions, revealing mechanisms like computational buffers and factual priming that improve accuracy.

Contribution

It uncovers the mechanisms by which reasoning improves factual recall in LLMs and proposes methods to enhance accuracy by managing hallucinations during reasoning.

Findings

Reasoning expands the model's knowledge retrieval capabilities.

Factual priming acts as a semantic bridge for better recall.

Hallucination risk increases with intermediate fact generation.

Abstract

While reasoning in LLMs plays a natural role in math, code generation, and multi-hop factual questions, its effect on simple, single-hop factual questions remains unclear. Such questions do not require step-by-step logical decomposition, making the utility of reasoning highly counterintuitive. Nevertheless, we find that enabling reasoning substantially expands the capability boundary of the model's parametric knowledge recall, unlocking correct answers that are otherwise effectively unreachable. Why does reasoning aid parametric knowledge recall when there are no complex reasoning steps to be done? To answer this, we design a series of hypothesis-driven controlled experiments, and identify two key driving mechanisms: (1) a computational buffer effect, where the model uses the generated reasoning tokens to perform latent computation independent of their semantic content; and (2) factual priming, where generating topically related facts acts as a semantic bridge that facilitates correct answer retrieval. Importantly, this latter generative self-retrieval mechanism carries inherent risks: we demonstrate that hallucinating intermediate facts during reasoning increases the likelihood of hallucinations in the final answer. Finally, we show that our insights can be harnessed to directly improve model accuracy by prioritizing reasoning trajectories that contain hallucination-free factual statements.