Entropy Distribution as a Fingerprint for Hallucinations in Generative Models

• URL: http://arxiv.org/abs/2605.28264v1
• Date: Wed, 27 May 2026 10:12:44 GMT
• Title: Entropy Distribution as a Fingerprint for Hallucinations in Generative Models
• Abstract summary: Calibrated Entropy Score (CES) is a lightweight algorithm for hallucination detection.<n>CES combines the mean signal with the maximum signal of the generated entropy through a calibrated reference CDF.<n>CES is statistically indistinguishable from multi-sample methods that require far greater computational cost.
• Score: 1.5135066879411019
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) often generate factually incorrect outputs, commonly termed hallucinations, that undermine trust and limit deployment in high-stakes settings. Existing hallucination detection methods typically require multiple forward passes, or access to model internals. In this work, we provide theoretical background and empirical evidence that the distribution of token-level entropies, beyond the mean captured by perplexity or length-normalised entropy, serves as a fingerprint of hallucination, with distributional shape and tail behaviour carrying independent signal. We formalize hallucination detection as a statistical hypothesis test and propose the Calibrated Entropy Score (CES), a lightweight algorithm requiring only a single forward pass and black-box access to token logits. CES combines the mean signal with the maximum signal of the generated entropy through a calibrated reference CDF, producing scores that are directly comparable across models and tasks. We establish finite-sample calibration guarantees via a novel random-length Dvoretzky--Kiefer--Wolfowitz inequality, and also prove that CES detects hallucinations with probability converging to one exponentially fast in the generation length. Across eight QA benchmarks and ten generator models spanning open-source and API access models, CES achieves the highest detection performance among all single-pass black-box methods while providing formal error guarantees that existing heuristics lack. Remarkably, CES is statistically indistinguishable from multi-sample methods that require far greater computational cost, closing the gap between lightweight and expensive detection and making it suitable for real-time, large-scale deployment.

Related papers

• TRE: Training-Free Hallucination Detection for Diffusion Language Models [9.622962795441852]
  We propose TRE, a training-free hallucination detection metric for D-LLMs.<n> TRE estimates hallucination risk directly from the entropy signals of a single generation, without requiring any detector training or repeated sampling.<n>Extensive experiments on multiple D-LLMs and QA datasets demonstrate that TRE achieves competitive performance, while enjoying strong generalizability, efficiency, and robustness.
  arXiv  Detail & Related papers  (2026-06-29T00:47:48Z)
• FLaG: Fine-Grained Latent Grouping for Hallucination Detection [22.6413030783939]
  Hallucinations in large language models (LLMs) arise from heterogeneous failure mechanisms, making reliable detection difficult for any single global uncertainty score.<n>We propose FLaG, a lightweight hallucination detection framework that models correctness through a set of latent evidence groups.<n>We show that FLaG consistently achieves SOTA performance, while exhibiting robust transfer across datasets and models, and remaining effective under limited supervision.
  arXiv  Detail & Related papers  (2026-05-29T19:30:39Z)
• EVEREST: An Evidential, Tail-Aware Transformer for Rare-Event Time-Series Forecasting [4.551615447454767]
  EVEREST is a transformer-based architecture for probabilistic rare-event forecasting.<n>It delivers calibrated predictions and tail-aware risk estimation.<n>It is applicable to high-stakes domains such as industrial monitoring, weather, and satellite diagnostics.
  arXiv  Detail & Related papers  (2026-01-26T23:15:20Z)
• Large Language Models Are Bad Dice Players: LLMs Struggle to Generate Random Numbers from Statistical Distributions [50.1404916337174]
  We present the first large-scale, statistically powered audit of native probabilistic sampling in large language models (LLMs)<n>We show that batch generation achieves only modest statistical validity, with a 13% median pass rate, while independent requests collapse almost entirely.<n>We conclude that current LLMs lack a functional internal sampler, necessitating the use of external tools for applications requiring statistical guarantees.
  arXiv  Detail & Related papers  (2026-01-08T22:33:12Z)
• Learning Auxiliary Tasks Improves Reference-Free Hallucination Detection in Open-Domain Long-Form Generation [78.78421340836915]
  We systematically investigate reference-free hallucination detection in open-domain long-form responses.<n>Our findings reveal that internal states are insufficient for reliably distinguishing between factual and hallucinated content.<n>We introduce a new paradigm, named RATE-FT, that augments fine-tuning with an auxiliary task for the model to jointly learn with the main task of hallucination detection.
  arXiv  Detail & Related papers  (2025-05-18T07:10:03Z)
• ShED-HD: A Shannon Entropy Distribution Framework for Lightweight Hallucination Detection on Edge Devices [2.193668764461763]
  Large Language Models (Ms.LS) have demonstrated impressive capabilities on a broad array of tasks, but their tendency to produce hallucinations.<n>We propose a novel hallucination detection framework that efficiently detects distinctive uncertainty patterns.
  arXiv  Detail & Related papers  (2025-03-23T23:47:26Z)
• Non-asymptotic bounds for forward processes in denoising diffusions: Ornstein-Uhlenbeck is hard to beat [49.1574468325115]
  This paper presents explicit non-asymptotic bounds on the forward diffusion error in total variation (TV)<n>We parametrise multi-modal data distributions in terms of the distance $R$ to their furthest modes and consider forward diffusions with additive and multiplicative noise.
  arXiv  Detail & Related papers  (2024-08-25T10:28:31Z)
• Data-Driven Abstractions via Binary-Tree Gaussian Processes for Formal Verification [0.22499166814992438]
  abstraction-based solutions based on Gaussian process (GP) regression have become popular for their ability to learn a representation of the latent system from data with a quantified error. We show that the binary-tree Gaussian process (BTGP) allows us to construct an interval Markov chain model of the unknown system. We provide a delocalized error quantification via a unified formula even when the true dynamics do not live in the function space of the BTGP.
  arXiv  Detail & Related papers  (2024-07-15T11:49:44Z)
• Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs [32.901839335074676]
  Hallucinations present a major challenge to the practical adoption of Large Language Models. Recent work by Farquhar et al. (2024) proposes semantic entropy (SE), which can detect hallucinations by estimating uncertainty in the space semantic meaning for a set of model generations. We propose SEPs, which directly approximate SE from the hidden states of a single generation.
  arXiv  Detail & Related papers  (2024-06-22T19:46:06Z)
• REAL Sampling: Boosting Factuality and Diversity of Open-Ended Generation via Asymptotic Entropy [93.8400683020273]
  Decoding methods for large language models (LLMs) usually struggle with the tradeoff between ensuring factuality and maintaining diversity. We propose REAL sampling, a decoding method that improved factuality and diversity over nucleus sampling.
  arXiv  Detail & Related papers  (2024-06-11T21:44:49Z)
• Prototype-based Aleatoric Uncertainty Quantification for Cross-modal Retrieval [139.21955930418815]
  Cross-modal Retrieval methods build similarity relations between vision and language modalities by jointly learning a common representation space. However, the predictions are often unreliable due to the Aleatoric uncertainty, which is induced by low-quality data, e.g., corrupt images, fast-paced videos, and non-detailed texts. We propose a novel Prototype-based Aleatoric Uncertainty Quantification (PAU) framework to provide trustworthy predictions by quantifying the uncertainty arisen from the inherent data ambiguity.
  arXiv  Detail & Related papers  (2023-09-29T09:41:19Z)
• FaDIn: Fast Discretized Inference for Hawkes Processes with General Parametric Kernels [82.53569355337586]
  This work offers an efficient solution to temporal point processes inference using general parametric kernels with finite support. The method's effectiveness is evaluated by modeling the occurrence of stimuli-induced patterns from brain signals recorded with magnetoencephalography (MEG) Results show that the proposed approach leads to an improved estimation of pattern latency than the state-of-the-art.
  arXiv  Detail & Related papers  (2022-10-10T12:35:02Z)
• SUOD: Accelerating Large-Scale Unsupervised Heterogeneous Outlier Detection [63.253850875265115]
  Outlier detection (OD) is a key machine learning (ML) task for identifying abnormal objects from general samples. We propose a modular acceleration system, called SUOD, to address it.
  arXiv  Detail & Related papers  (2020-03-11T00:22:50Z)

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.