Detection and Mitigation of Hallucination in Large Reasoning Models: A Mechanistic Perspective

• URL: http://arxiv.org/abs/2505.12886v1
• Date: Mon, 19 May 2025 09:16:40 GMT
• Title: Detection and Mitigation of Hallucination in Large Reasoning Models: A Mechanistic Perspective
• Authors: Zhongxiang Sun, Qipeng Wang, Haoyu Wang, Xiao Zhang, Jun Xu,
• Abstract summary: Reasoning Hallucinations are logically coherent but factually incorrect reasoning traces.<n>These errors are embedded within structured reasoning, making them more difficult to detect and potentially more harmful.<n>We propose the Reasoning Score, which quantifies the depth of reasoning by measuring the divergence between logits.<n>We also introduce GRPO-R, an enhanced reinforcement learning algorithm that incorporates step-level deep reasoning rewards via potential-based shaping.
• Score: 11.013059864022667
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Reasoning Models (LRMs) have shown impressive capabilities in multi-step reasoning tasks. However, alongside these successes, a more deceptive form of model error has emerged--Reasoning Hallucination--where logically coherent but factually incorrect reasoning traces lead to persuasive yet faulty conclusions. Unlike traditional hallucinations, these errors are embedded within structured reasoning, making them more difficult to detect and potentially more harmful. In this work, we investigate reasoning hallucinations from a mechanistic perspective. We propose the Reasoning Score, which quantifies the depth of reasoning by measuring the divergence between logits obtained from projecting late layers of LRMs to the vocabulary space, effectively distinguishing shallow pattern-matching from genuine deep reasoning. Using this score, we conduct an in-depth analysis on the ReTruthQA dataset and identify two key reasoning hallucination patterns: early-stage fluctuation in reasoning depth and incorrect backtracking to flawed prior steps. These insights motivate our Reasoning Hallucination Detection (RHD) framework, which achieves state-of-the-art performance across multiple domains. To mitigate reasoning hallucinations, we further introduce GRPO-R, an enhanced reinforcement learning algorithm that incorporates step-level deep reasoning rewards via potential-based shaping. Our theoretical analysis establishes stronger generalization guarantees, and experiments demonstrate improved reasoning quality and reduced hallucination rates.

Related papers

• Lost at the Beginning of Reasoning [82.18834329384514]
  We show that the first reasoning step exerts a disproportionately large influence on the final prediction.<n>We propose an efficient sampling strategy that leverages a reward model to identify and retain high-quality first reasoning steps.<n>We introduce a new benchmark specifically constructed with deliberately flawed first reasoning steps to systematically evaluate model self-correction capabilities.
  arXiv  Detail & Related papers  (2025-06-27T09:53:57Z)
• CLATTER: Comprehensive Entailment Reasoning for Hallucination Detection [60.98964268961243]
  We propose that guiding models to perform a systematic and comprehensive reasoning process allows models to execute much finer-grained and accurate entailment decisions.<n>We define a 3-step reasoning process, consisting of (i) claim decomposition, (ii) sub-claim attribution and entailment classification, and (iii) aggregated classification, showing that such guided reasoning indeed yields improved hallucination detection.
  arXiv  Detail & Related papers  (2025-06-05T17:02:52Z)
• Joint Evaluation of Answer and Reasoning Consistency for Hallucination Detection in Large Reasoning Models [12.270274049887298]
  Reasoning traces can be redundant or logically inconsistent, making them a new source of hallucination.<n>Existing hallucination detection methods focus primarily on answer-level uncertainty.<n>We propose RACE, a novel framework specifically tailored for hallucination detection in LRMs.
  arXiv  Detail & Related papers  (2025-06-05T09:54:04Z)
• The Hallucination Dilemma: Factuality-Aware Reinforcement Learning for Large Reasoning Models [63.98194996746229]
  Large language models (LLMs) have significantly advanced in reasoning tasks through reinforcement learning (RL) optimization.<n>However, reasoning-oriented RL fine-tuning significantly increases the prevalence of hallucinations.<n>We propose Factuality-aware Step-wise Policy Optimization (FSPO), an innovative RL fine-tuning algorithm incorporating explicit factuality verification.
  arXiv  Detail & Related papers  (2025-05-30T14:23:32Z)
• MIRAGE: Assessing Hallucination in Multimodal Reasoning Chains of MLLM [58.2298313720146]
  Multimodal hallucinations are multi-sourced and arise from diverse causes.<n>Existing benchmarks fail to adequately distinguish between perception-induced hallucinations and reasoning-induced hallucinations.
  arXiv  Detail & Related papers  (2025-05-30T05:54:36Z)
• Are Reasoning Models More Prone to Hallucination? [70.04436965009072]
  Recently evolved large reasoning models (LRMs) show powerful performance in solving complex tasks with long chain-of-thought (CoT) reasoning capability.<n>Are reasoning models more prone to hallucination?<n>This paper addresses the question from three perspectives.
  arXiv  Detail & Related papers  (2025-05-29T16:53:41Z)
• More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models [43.465268635499754]
  Test-time compute has empowered large language models to generate extended reasoning chains.<n>As generations become longer, models tend to drift away from image-grounded content and rely more heavily on language priors.
  arXiv  Detail & Related papers  (2025-05-23T05:08:40Z)
• Auditing Meta-Cognitive Hallucinations in Reasoning Large Language Models [8.97308732968526]
  We study the causality of hallucinations under constrained knowledge domains by auditing the Chain-of-Thought trajectory.<n>Our analysis reveals that in long-CoT settings, RLLMs can iteratively reinforce biases and errors through flawed reflective reasoning.<n>Surprisingly, even direct interventions at the origin of hallucinations often fail to reverse their effects.
  arXiv  Detail & Related papers  (2025-05-19T14:11:09Z)
• Why and How LLMs Hallucinate: Connecting the Dots with Subsequence Associations [82.42811602081692]
  This paper introduces a subsequence association framework to systematically trace and understand hallucinations.<n>Key insight is hallucinations that arise when dominant hallucinatory associations outweigh faithful ones.<n>We propose a tracing algorithm that identifies causal subsequences by analyzing hallucination probabilities across randomized input contexts.
  arXiv  Detail & Related papers  (2025-04-17T06:34:45Z)
• Delusions of Large Language Models [62.43923767408462]
  Large Language Models often generate factually incorrect but plausible outputs, known as hallucinations.<n>We identify a more insidious phenomenon, LLM delusion, defined as high belief hallucinations, incorrect outputs with abnormally high confidence, making them harder to detect and mitigate.
  arXiv  Detail & Related papers  (2025-03-09T17:59:16Z)
• Causality can systematically address the monsters under the bench(marks) [64.36592889550431]
  Benchmarks are plagued by various biases, artifacts, or leakage.<n>Models may behave unreliably due to poorly explored failure modes.<n> causality offers an ideal framework to systematically address these challenges.
  arXiv  Detail & Related papers  (2025-02-07T17:01:37Z)
• A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences [5.141416267381492]
  We consider the case of syllogistic reasoning, an area of deductive reasoning studied extensively in logic and cognitive psychology. We investigate the effects of chain-of-thought reasoning, in-context learning, and supervised fine-tuning on syllogistic reasoning. Our results suggest that the behavior of pre-trained LLMs can be explained by cognitive science.
  arXiv  Detail & Related papers  (2024-06-17T08:59:04Z)
• In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigation [36.31646727970656]
  Large language models (LLMs) frequently hallucinate and produce factual errors. correct generations tend to have sharper context activations in the hidden states of the in-context tokens, compared to the incorrect ones. We propose an entropy-based metric to quantify the sharpness'' among the in-context hidden states and incorporate it into the decoding process.
  arXiv  Detail & Related papers  (2024-03-03T15:53:41Z)
• 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)
• A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning [73.77088902676306]
  We take a closer look at the self-verification abilities of large language models (LLMs) in the context of logical reasoning. Our main findings suggest that existing LLMs could struggle to identify fallacious reasoning steps accurately and may fall short of guaranteeing the validity of self-verification methods.
  arXiv  Detail & Related papers  (2023-11-14T07:13:10Z)

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.