Shadows in the Attention: Contextual Perturbation and Representation Drift in the Dynamics of Hallucination in LLMs

• URL: http://arxiv.org/abs/2505.16894v1
• Date: Thu, 22 May 2025 16:50:58 GMT
• Title: Shadows in the Attention: Contextual Perturbation and Representation Drift in the Dynamics of Hallucination in LLMs
• Authors: Zeyu Wei, Shuo Wang, Xiaohui Rong, Xuemin Liu, He Li,
• Abstract summary: We present the first systematic study linking hallucination incidence to internal-state drift induced by context injection.<n>Using TruthfulQA, we construct two 16-round "titration" tracks per question.<n>We track overt hallucination rates with a tri-perspective detector and covert dynamics via cosine, entropy, JS and Spearman drifts of hidden states and attention maps.
• Score: 6.190663515080656
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Hallucinations -- plausible yet erroneous outputs -- remain a critical barrier to reliable deployment of large language models (LLMs). We present the first systematic study linking hallucination incidence to internal-state drift induced by incremental context injection. Using TruthfulQA, we construct two 16-round "titration" tracks per question: one appends relevant but partially flawed snippets, the other injects deliberately misleading content. Across six open-source LLMs, we track overt hallucination rates with a tri-perspective detector and covert dynamics via cosine, entropy, JS and Spearman drifts of hidden states and attention maps. Results reveal (1) monotonic growth of hallucination frequency and representation drift that plateaus after 5--7 rounds; (2) relevant context drives deeper semantic assimilation, producing high-confidence "self-consistent" hallucinations, whereas irrelevant context induces topic-drift errors anchored by attention re-routing; and (3) convergence of JS-Drift ($\sim0.69$) and Spearman-Drift ($\sim0$) marks an "attention-locking" threshold beyond which hallucinations solidify and become resistant to correction. Correlation analyses expose a seesaw between assimilation capacity and attention diffusion, clarifying size-dependent error modes. These findings supply empirical foundations for intrinsic hallucination prediction and context-aware mitigation mechanisms.

Related papers

• Causally-Grounded Dual-Path Attention Intervention for Object Hallucination Mitigation in LVLMs [26.144870818163387]
  We propose a framework that models hallucination process via a structural causal graph.<n>We introduce VTACR, a novel metric that quantifies the modality contribution imbalance during decoding.<n>We design a fine-language attention intervention mechanism that dynamically adjusts token- and layer-wise attention.
  arXiv  Detail & Related papers  (2025-11-12T06:13:26Z)
• Distributional Semantics Tracing: A Framework for Explaining Hallucinations in Large Language Models [4.946483489399819]
  Large Language Models (LLMs) are prone to hallucination, the generation of factually incorrect statements.<n>This work investigates the intrinsic, architectural origins of this failure mode through three primary contributions.
  arXiv  Detail & Related papers  (2025-10-07T16:40:31Z)
• How Large Language Models are Designed to Hallucinate [0.42970700836450487]
  We argue that hallucination is a structural outcome of the transformer architecture.<n>Our contribution is threefold: (1) a comparative account showing why existing explanations are insufficient; (2) a predictive taxonomy of hallucination linked to existential structures with proposed benchmarks; and (3) design directions toward "truth-constrained" architectures capable of withholding or deferring when disclosure is absent.
  arXiv  Detail & Related papers  (2025-09-19T16:46:27Z)
• Two Causes, Not One: Rethinking Omission and Fabrication Hallucinations in MLLMs [31.601057368065877]
  Existing methods, based on the flawed assumption that omission and fabrication hallucinations share a common cause, often reduce omissions only to trigger more fabrications.<n>In this work, we overturn this view by demonstrating that omission hallucinations arise from insufficient confidence when mapping perceived visual features to linguistic expressions.<n>We propose the Visual-Semantic Attention Potential Field, a conceptual framework that reveals how visual evidence to infer the presence or absence of objects.
  arXiv  Detail & Related papers  (2025-08-30T05:47:41Z)
• SHALE: A Scalable Benchmark for Fine-grained Hallucination Evaluation in LVLMs [52.03164192840023]
  Large Vision-Language Models (LVLMs) still suffer from hallucinations, i.e., generating content inconsistent with input or established world knowledge.<n>We propose an automated data construction pipeline that produces scalable, controllable, and diverse evaluation data.<n>We construct SHALE, a benchmark designed to assess both faithfulness and factuality hallucinations.
  arXiv  Detail & Related papers  (2025-08-13T07:58:01Z)
• MIHBench: Benchmarking and Mitigating Multi-Image Hallucinations in Multimodal Large Language Models [73.20126092411776]
  We conduct the first systematic study of hallucinations in multi-image MLLMs.<n>We propose MIHBench, a benchmark specifically tailored for evaluating object-related hallucinations across multiple images.<n>MIHBench comprises three core tasks: Multi-Image Object Existence Hallucination, Multi-Image Object Count Hallucination, and Object Identity Consistency Hallucination.
  arXiv  Detail & Related papers  (2025-08-01T15:49:29Z)
• Chain-of-Thought Prompting Obscures Hallucination Cues in Large Language Models: An Empirical Evaluation [4.265468175793552]
  Chain-of-Thought (CoT) prompting can mitigate hallucinations by encouraging step-by-step reasoning.<n>Our study highlights an overlooked trade-off in the use of reasoning.
  arXiv  Detail & Related papers  (2025-06-20T15:49:37Z)
• When Semantics Mislead Vision: Mitigating Large Multimodal Models Hallucinations in Scene Text Spotting and Understanding [72.15848305976706]
  Large Multimodal Models (LMMs) have achieved impressive progress in visual perception and reasoning.<n>When confronted with visually ambiguous or non-semantic scene text, they often struggle to accurately spot and understand the content.<n>We propose a training-free semantic hallucination mitigation framework comprising two key components.
  arXiv  Detail & Related papers  (2025-06-05T19:53:19Z)
• 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)
• 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)
• HalluLens: LLM Hallucination Benchmark [49.170128733508335]
  Large language models (LLMs) often generate responses that deviate from user input or training data, a phenomenon known as "hallucination"<n>This paper introduces a comprehensive hallucination benchmark, incorporating both new extrinsic and existing intrinsic evaluation tasks.
  arXiv  Detail & Related papers  (2025-04-24T13:40:27Z)
• 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)
• On Large Language Models' Hallucination with Regard to Known Facts [74.96789694959894]
  Large language models are successful in answering factoid questions but are also prone to hallucination. We investigate the phenomenon of LLMs possessing correct answer knowledge yet still hallucinating from the perspective of inference dynamics. Our study shed light on understanding the reasons for LLMs' hallucinations on their known facts, and more importantly, on accurately predicting when they are hallucinating.
  arXiv  Detail & Related papers  (2024-03-29T06:48:30Z)
• 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)
• Quantity Matters: Towards Assessing and Mitigating Number Hallucination in Large Vision-Language Models [57.42800112251644]
  We focus on a specific type of hallucination-number hallucination, referring to models incorrectly identifying the number of certain objects in pictures. We devise a training approach aimed at improving consistency to reduce number hallucinations, which leads to an 8% enhancement in performance over direct finetuning methods.
  arXiv  Detail & Related papers  (2024-03-03T02:31:11Z)
• Retrieve Only When It Needs: Adaptive Retrieval Augmentation for Hallucination Mitigation in Large Language Models [68.91592125175787]
  Hallucinations pose a significant challenge for the practical implementation of large language models (LLMs) We present Rowen, a novel approach that enhances LLMs with a selective retrieval augmentation process tailored to address hallucinations.
  arXiv  Detail & Related papers  (2024-02-16T11:55:40Z)
• The Troubling Emergence of Hallucination in Large Language Models -- An Extensive Definition, Quantification, and Prescriptive Remediations [10.20632187568563]
  We offer a discourse on profiling hallucination based on its degree, orientation, and category. We categorize hallucination into six types: (i) acronym ambiguity, (ii) numeric nuisance, (iii) generated golem, (iv) virtual voice, (v) geographic erratum, and (vi) time wrap. We propose Hallucination Vulnerability Index (HVI) as a tool for the wider NLP community, with the potential to serve as a rubric in AI-related policy-making.
  arXiv  Detail & Related papers  (2023-10-08T03:31:29Z)

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.