Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs

• URL: http://arxiv.org/abs/2603.09906v1
• Date: Tue, 10 Mar 2026 16:59:20 GMT
• Title: Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
• Abstract summary: We find that enabling reasoning substantially expands the capability boundary of the model's parametric knowledge recall.<n>We show that our insights can be harnessed to directly improve model accuracy by prioritizing reasoning trajectories that contain hallucination-free factual statements.
• Score: 45.274171760492685
• License: http://creativecommons.org/licenses/by/4.0/
• 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.

Related papers

• CausalFlip: A Benchmark for LLM Causal Judgment Beyond Semantic Matching [50.65932158912512]
  We propose a new causal reasoning benchmark, CausalFlip, to encourage the development of new large language models.<n>CaulFlip consists of causal judgment questions built over event triples that could form different confounder, chain, and collider relations.<n>We evaluate LLMs under multiple training paradigms, including answer-only training, explicit Chain-of-Thought supervision, and a proposed internalized causal reasoning approach.
  arXiv  Detail & Related papers  (2026-02-23T18:06:15Z)
• Towards Generalizable Reasoning: Group Causal Counterfactual Policy Optimization for LLM Reasoning [50.352417879912515]
  Large language models (LLMs) excel at complex tasks with advances in reasoning capabilities.<n>We propose Group Causal Counterfactual Policy Optimization to explicitly train LLMs to learn generalizable reasoning patterns.<n>We then construct token-level advantages from this reward and optimize the policy, encouraging LLMs to favor reasoning patterns that are process-valid and counterfactually robust.
  arXiv  Detail & Related papers  (2026-02-06T08:03:11Z)
• RAVR: Reference-Answer-guided Variational Reasoning for Large Language Models [21.671577399379885]
  We introduce RAVR, an end-to-end framework that uses answer-conditioned reasoning as a variational surrogate for question-only reasoning.<n> RAVR reduces hesitation, strengthens conclusion consolidation, and promotes problem-specific strategies in reasoning.
  arXiv  Detail & Related papers  (2025-10-29T06:18:37Z)
• Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models [15.797612515648412]
  Large reasoning models (LRMs) exhibit unprecedented capabilities in solving complex problems through Chain-of-Thought (CoT) reasoning.<n>Recent studies reveal that their final answers often contradict their own reasoning traces.<n>We hypothesize that this inconsistency stems from two competing mechanisms for generating answers: CoT reasoning and memory retrieval.<n>We introduce FARL, a novel fine-tuning framework that integrates memory unlearning with reinforcement learning.
  arXiv  Detail & Related papers  (2025-09-29T01:13:33Z)
• On the Fundamental Impossibility of Hallucination Control in Large Language Models [0.0]
  Impossibility Theorem: no LLM performing non-trivial knowledge aggregation can simultaneously achieve truthful knowledge representation, semantic information conservation, and revelation of relevant knowledge.<n>We prove this by modeling inference as an auction of ideas, where distributed components compete to influence responses using encoded knowledge.<n>We show that hallucination and imagination are mathematically identical, and both violate at least one of the four essential properties.
  arXiv  Detail & Related papers  (2025-06-04T23:28:39Z)
• Disentangling Memory and Reasoning Ability in Large Language Models [97.26827060106581]
  We propose a new inference paradigm that decomposes the complex inference process into two distinct and clear actions.<n>Our experiment results show that this decomposition improves model performance and enhances the interpretability of the inference process.
  arXiv  Detail & Related papers  (2024-11-20T17:55:38Z)
• Failure Modes of LLMs for Causal Reasoning on Narratives [51.19592551510628]
  We investigate the interaction between world knowledge and logical reasoning.<n>We find that state-of-the-art large language models (LLMs) often rely on superficial generalizations.<n>We show that simple reformulations of the task can elicit more robust reasoning behavior.
  arXiv  Detail & Related papers  (2024-10-31T12:48:58Z)
• Distilling Reasoning Ability from Large Language Models with Adaptive Thinking [54.047761094420174]
  Chain of thought finetuning (cot-finetuning) aims to endow small language models (SLM) with reasoning ability to improve their performance towards specific tasks.<n>Most existing cot-finetuning methods adopt a pre-thinking mechanism, allowing the SLM to generate a rationale before providing an answer.<n>This mechanism enables SLM to analyze and think about complex questions, but it also makes answer correctness highly sensitive to minor errors in rationale.<n>We propose a robust post-thinking mechanism to generate answers before rationale.
  arXiv  Detail & Related papers  (2024-04-14T07:19:27Z)
• Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation [110.71955853831707]
  We view LMs as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths.
  arXiv  Detail & Related papers  (2024-02-05T18:25:51Z)
• Boosting Language Models Reasoning with Chain-of-Knowledge Prompting [18.326858925174605]
  Chain-of-Knowledge (CoK) prompting aims at eliciting explicit pieces of knowledge evidence in the form of structure triple. Benefiting from CoK, we additionally introduce a F2-Verification method to estimate the reliability of the reasoning chains. Extensive experiments demonstrate that our method can further improve the performance of commonsense, factual, symbolic, and arithmetic reasoning tasks.
  arXiv  Detail & Related papers  (2023-06-10T12:42:36Z)
• Faithful Reasoning Using Large Language Models [12.132449274592668]
  We show how LMs can be made to perform faithful multi-step reasoning via a process whose causal structure mirrors the underlying logical structure of the problem. Our approach works by chaining together reasoning steps, where each step results from calls to two fine-tuned LMs. We demonstrate the effectiveness of our model on multi-step logical deduction and scientific question-answering, showing that it outperforms baselines on final answer accuracy.
  arXiv  Detail & Related papers  (2022-08-30T13:44:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.