D$^2$HScore: Reasoning-Aware Hallucination Detection via Semantic Breadth and Depth Analysis in LLMs

• URL: http://arxiv.org/abs/2509.11569v1
• Date: Mon, 15 Sep 2025 04:28:38 GMT
• Title: D$^2$HScore: Reasoning-Aware Hallucination Detection via Semantic Breadth and Depth Analysis in LLMs
• Authors: Yue Ding, Xiaofang Zhu, Tianze Xia, Junfei Wu, Xinlong Chen, Qiang Liu, Liang Wang,
• Abstract summary: This work revisits hallucination detection from the perspective of model architecture and generation dynamics.<n>We propose textbfD$2$HScore (Dispersion and Drift-based Hallucination Score), a training-free and label-free framework.<n>Experiments across five open-source Language Models and five widely used benchmarks demonstrate that D$2$HScore consistently outperforms existing training-free baselines.
• Score: 15.665202830841046
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Although large Language Models (LLMs) have achieved remarkable success, their practical application is often hindered by the generation of non-factual content, which is called "hallucination". Ensuring the reliability of LLMs' outputs is a critical challenge, particularly in high-stakes domains such as finance, security, and healthcare. In this work, we revisit hallucination detection from the perspective of model architecture and generation dynamics. Leveraging the multi-layer structure and autoregressive decoding process of LLMs, we decompose hallucination signals into two complementary dimensions: the semantic breadth of token representations within each layer, and the semantic depth of core concepts as they evolve across layers. Based on this insight, we propose \textbf{D$^2$HScore (Dispersion and Drift-based Hallucination Score)}, a training-free and label-free framework that jointly measures: (1) \textbf{Intra-Layer Dispersion}, which quantifies the semantic diversity of token representations within each layer; and (2) \textbf{Inter-Layer Drift}, which tracks the progressive transformation of key token representations across layers. To ensure drift reflects the evolution of meaningful semantics rather than noisy or redundant tokens, we guide token selection using attention signals. By capturing both the horizontal and vertical dynamics of representation during inference, D$^2$HScore provides an interpretable and lightweight proxy for hallucination detection. Extensive experiments across five open-source LLMs and five widely used benchmarks demonstrate that D$^2$HScore consistently outperforms existing training-free baselines.

Related papers

• Hallucination Begins Where Saliency Drops [18.189047289404325]
  hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token.<n>We introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token.<n>Our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution.
  arXiv  Detail & Related papers  (2026-01-28T05:50:52Z)
• Revealing Perception and Generation Dynamics in LVLMs: Mitigating Hallucinations via Validated Dominance Correction [59.801614364841775]
  Large Vision-Language Models (LVLMs) have shown remarkable capabilities, yet hallucinations remain a persistent challenge.<n>This work presents a systematic analysis of the internal evolution of visual perception and token generation in LVLMs.<n>We devise the VDC (d Dominance Correction) strategy, which detects unsupported tokens and replaces them with validated ones to improve output reliability.
  arXiv  Detail & Related papers  (2025-12-21T17:05:42Z)
• HalluShift++: Bridging Language and Vision through Internal Representation Shifts for Hierarchical Hallucinations in MLLMs [17.130698952440316]
  hallucinations in large language models (MLLMs) have become increasingly crucial in the model development process.<n>We propose the hypothesis that hallucination manifests as measurable irregularities within the internal layer dynamics of MLLMs.<n>By incorporating such modifications, textsctextscHalluShift++ broadens the efficacy of hallucination detection from text-based large language models.
  arXiv  Detail & Related papers  (2025-12-08T16:24:46Z)
• Gradient-Attention Guided Dual-Masking Synergetic Framework for Robust Text-based Person Retrieval [15.126709823382539]
  This work advances Contrastive Language-Image Pre-training (CLIP) for person representation learning.<n>We develop a noise-resistant data construction pipeline that leverages the in-context learning capabilities of MLLMs.<n>We introduce the GA-DMS framework, which improves cross-modal alignment by adaptively masking noisy textual tokens.
  arXiv  Detail & Related papers  (2025-09-11T03:06:22Z)
• Beyond ROUGE: N-Gram Subspace Features for LLM Hallucination Detection [5.0106565473767075]
  Large Language Models (LLMs) have demonstrated effectiveness across a wide variety of tasks involving natural language.<n>A fundamental problem of hallucinations still plagues these models, limiting their trustworthiness in generating consistent, truthful information.<n>We propose a novel approach inspired by ROUGE that constructs an N-Gram frequency tensor from LLM-generated text.<n>This tensor captures richer semantic structure by encoding co-occurrence patterns, enabling better differentiation between factual and hallucinated content.
  arXiv  Detail & Related papers  (2025-09-03T18:52:24Z)
• How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer Decoding [39.342366994703376]
  We introduce a probing framework to analyze how MLLMs process visual and textual inputs across layers.<n>We show that while the overall stage-wise structure remains stable across variations in visual tokenization, instruction tuning data, and pretraining corpus, the specific layer allocation to each stage shifts.
  arXiv  Detail & Related papers  (2025-08-27T21:22:01Z)
• LISA: A Layer-wise Integration and Suppression Approach for Hallucination Mitigation in Multimodal Large Language Models [8.122679857175315]
  Multimodal Large Language Models (MLLMs) excel in vision-language tasks but remain prone to object hallucinations.<n>We propose textbfLISA, which enhances generation consistency through hierarchical modulation and multi-layer fusion.<n>Experiments show that LISA reduces hallucinations by up to 53.6% in $mathrmCHAIR_I$ and improves POPE F1 by 4.5%.
  arXiv  Detail & Related papers  (2025-07-25T09:48:23Z)
• Empowering Sparse-Input Neural Radiance Fields with Dual-Level Semantic Guidance from Dense Novel Views [66.1245505423179]
  We show that rendered semantics can be treated as a more robust form of augmented data than rendered RGB.<n>Our method enhances NeRF's performance by incorporating guidance derived from the rendered semantics.
  arXiv  Detail & Related papers  (2025-03-04T03:13:44Z)
• Towards Semantic Equivalence of Tokenization in Multimodal LLM [149.11720372278273]
  Vision tokenization is essential for semantic alignment between vision and language.<n>This paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok)<n>SeTok groups visual features into semantic units via a dynamic clustering algorithm.<n>The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features.
  arXiv  Detail & Related papers  (2024-06-07T17:55:43Z)
• Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference [59.91176945361035]
  We introduce Visual Tokens Withdrawal (VTW), a plug-and-play module to boost MLLMs for rapid inference.<n>VTW strategically withdraws vision tokens at a certain layer, enabling only text tokens to engage in subsequent layers.<n>Our approach can cut computational overhead by over 40% across diverse multimodal tasks while maintaining performance.
  arXiv  Detail & Related papers  (2024-05-09T14:38:53Z)
• LEAF: Unveiling Two Sides of the Same Coin in Semi-supervised Facial Expression Recognition [56.22672276092373]
  Semi-supervised learning has emerged as a promising approach to tackle the challenge of label scarcity in facial expression recognition.<n>We propose a unified framework termed hierarchicaL dEcoupling And Fusing (LEAF) to coordinate expression-relevant representations and pseudo-labels for semi-supervised FER.
  arXiv  Detail & Related papers  (2024-04-23T13:43:33Z)
• LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge [58.82222646803248]
  Multimodal Large Language Models (MLLMs) have endowed LLMs with the ability to perceive and understand multi-modal signals. Most of the existing MLLMs mainly adopt vision encoders pretrained on coarsely aligned image-text pairs, leading to insufficient extraction and reasoning of visual knowledge. We propose a dual-Level vIsual knedgeOwl eNhanced Multimodal Large Language Model (LION), which empowers the MLLM by injecting visual knowledge in two levels.
  arXiv  Detail & Related papers  (2023-11-20T15:56:44Z)

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.