The Phenomenology of Hallucinations

• URL: http://arxiv.org/abs/2603.13911v1
• Date: Sat, 14 Mar 2026 11:55:55 GMT
• Title: The Phenomenology of Hallucinations
• Abstract summary: We show that language models hallucinate not because they fail to detect uncertainty, but because of a failure to integrate it into output generation.<n>Across architectures, uncertain inputs are reliably identified, occupying high-dimensional regions with 2-3$times$ the intrinsic dimensionality of factual inputs.<n>However, this internal signal is weakly coupled to the output layer: uncertainty migrates into low-sensitivity subspaces, becoming geometrically amplified yet functionally silent.
• Score: 2.5782420501870296
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We show that language models hallucinate not because they fail to detect uncertainty, but because of a failure to integrate it into output generation. Across architectures, uncertain inputs are reliably identified, occupying high-dimensional regions with 2-3$\times$ the intrinsic dimensionality of factual inputs. However, this internal signal is weakly coupled to the output layer: uncertainty migrates into low-sensitivity subspaces, becoming geometrically amplified yet functionally silent. Topological analysis shows that uncertainty representations fragment rather than converging to a unified abstention state, while gradient and Fisher probes reveal collapsing sensitivity along the uncertainty direction. Because cross-entropy training provides no attractor for abstention and uniformly rewards confident prediction, associative mechanisms amplify these fractured activations until residual coupling forces a committed output despite internal detection. Causal interventions confirm this account by restoring refusal when uncertainty is directly connected to logits.

Related papers

• Beyond Boundary Noise: Aggregated Aleatoric Uncertainty Fails to Capture Presence Ambiguity in 3D Lung Nodule Segmentation [13.944682026264472]
  We systematically evaluate how well aleatoric uncertainty captures presence ambiguity in 3D lung segmentation.<n>We find that entropy-based uncertainty maps align with boundary noise but carry insufficient discriminative signal for presence ambiguity.<n>A lightweight supervised ambiguity head trained on frozen segmentation features substantially outperforms entropy-aggregation-based baselines.
  arXiv  Detail & Related papers  (2026-08-14T12:12:34Z)
• Geometric Collapse: When Vision Models Fail to Verify Physical Causality [52.43819179934414]
  We propose Scrambled Edges, a controlled counterfactual that injects salient edge-like cues.<n>With energy-matched and structure-matched controls, we isolate the effect of unsupported edge evidence from high-frequency energy and edge sparsity.<n>Across CNN/ViT/SSL depth predictors, Scrambled Edges induce up to 3.2x larger deviation from clean predictions than energy-matched noise.
  arXiv  Detail & Related papers  (2026-07-08T00:04:02Z)
• Sanity Checks for Long-Form Hallucination Detection [6.501493411588847]
  Hallucination detection methods for large language models operate on chain-of-thought reasoning traces, yet it remains unclear whether they evaluate the reasoning itself or merely exploit surface correlates of the final answer.<n>We introduce a controlled-invariance methodology that exposes this distinction through two oracle tests: textscForce, which replaces each response's final answer with the ground truth while preserving the reasoning trace, and textscRemove, which strips answer-announcement steps while leaving the trajectory intact.<n>This reveals if their predictive power derives from answer-level artifacts rather than from the structure
  arXiv  Detail & Related papers  (2026-05-08T18:00:58Z)
• UnGAP: Uncertainty-Guided Affine Prompting for Real-Time Crack Segmentation [16.902729686979303]
  Uncertainty-Prompted Feature Modulator (UPFM) treats aleatoric uncertainty as an active visual prompt rather than a mere output.<n>UnGAP balances superior segmentation accuracy with real-time inference speed.
  arXiv  Detail & Related papers  (2026-05-04T09:18:57Z)
• Neural Uncertainty Principle: A Unified View of Adversarial Fragility and LLM Hallucination [60.197429875410286]
  Adversarial vulnerability in vision and hallucination in large language models are conventionally viewed as separate problems.<n>Formalizing a Neural Uncertainty Principle (NUP) under a loss-induced state, we find that in near-bound regimes, further compression must be accompanied by increased sensitivity dispersion.<n>In vision, masking highly coupled components improves robustness without costly adversarial training.<n>In language, the same prefill-stage probe detects hallucination risk before generating any answer tokens.
  arXiv  Detail & Related papers  (2026-03-20T02:07:10Z)
• Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models [62.932580559941414]
  Vision-Language Models (VLMs) frequently "hallucinate" - generate plausible yet factually incorrect statements.<n>We propose a new paradigm for diagnosing hallucinations, recasting them from static output errors into dynamic pathologies of a model's computational cognition.
  arXiv  Detail & Related papers  (2026-03-16T17:20:38Z)
• On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation [51.56484100374058]
  We formalise such information-conditioned interaction patterns as behavioural dependency.<n>This induces a probe-relative notion of $$-behavioural equivalence and a within-policy behavioural distance.<n>Results identify structural conditions under which probe-conditioned behavioural separation is not preserved under common policy transformations.
  arXiv  Detail & Related papers  (2026-02-24T22:55:21Z)
• Self-Aware Object Detection via Degradation Manifolds [3.8265249634979734]
  In safety-critical settings, it is insufficient to produce predictions without assessing whether the input remains within the detector's nominal operating regime.<n>We introduce a degradation-aware self-awareness framework based on degradation manifold.<n>Our method augments a standard detection backbone with a lightweight embedding head trained via contrastive learning.
  arXiv  Detail & Related papers  (2026-02-20T17:58:46Z)
• Agentic Uncertainty Quantification [76.94013626702183]
  We propose a unified Dual-Process Agentic UQ (AUQ) framework that transforms verbalized uncertainty into active, bi-directional control signals.<n>Our architecture comprises two complementary mechanisms: System 1 (Uncertainty-Aware Memory, UAM), which implicitly propagates verbalized confidence and semantic explanations to prevent blind decision-making; and System 2 (Uncertainty-Aware Reflection, UAR), which utilizes these explanations as rational cues to trigger targeted inference-time resolution only when necessary.
  arXiv  Detail & Related papers  (2026-01-22T07:16:26Z)
• HaluNet: Multi-Granular Uncertainty Modeling for Efficient Hallucination Detection in LLM Question Answering [12.183015986299438]
  We present textbfHaluNet, a lightweight and trainable neural framework that integrates multi granular token level uncertainties.<n> Experiments on SQuAD, TriviaQA, and Natural Questions show that HaluNet delivers strong detection performance and favorable computational efficiency.
  arXiv  Detail & Related papers  (2025-12-31T02:03:10Z)
• Semantic Energy: Detecting LLM Hallucination Beyond Entropy [106.92072182161712]
  Large Language Models (LLMs) are being increasingly deployed in real-world applications, but they remain susceptible to hallucinations.<n>Uncertainty estimation is a feasible approach to detect such hallucinations.<n>We introduce Semantic Energy, a novel uncertainty estimation framework.
  arXiv  Detail & Related papers  (2025-08-20T07:33:50Z)
• RePPL: Recalibrating Perplexity by Uncertainty in Semantic Propagation and Language Generation for Explainable QA Hallucination Detection [26.186204911845866]
  hallucinations remain a vital obstacle to large language models' trustworthy use.<n>We propose RePPL to recalibrate uncertainty measurement by these two aspects.<n>Our method achieves the best comprehensive detection performance across various QA datasets.
  arXiv  Detail & Related papers  (2025-05-21T11:23:05Z)
• Investigating the Impact of Model Instability on Explanations and Uncertainty [43.254616360807496]
  We simulate uncertainty in text input by introducing noise at inference time. We find that high uncertainty doesn't necessarily imply low explanation plausibility. This suggests that noise-augmented models may be better at identifying salient tokens when uncertain.
  arXiv  Detail & Related papers  (2024-02-20T13:41:21Z)
• The Hidden Uncertainty in a Neural Networks Activations [105.4223982696279]
  The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution correlates with a model's epistemic uncertainty, thus indicating its ability to generalise to novel inputs.
  arXiv  Detail & Related papers  (2020-12-05T17:30:35Z)

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.