Learned Hallucination Detection in Black-Box LLMs using Token-level Entropy Production Rate

• URL: http://arxiv.org/abs/2509.04492v1
• Date: Mon, 01 Sep 2025 13:34:21 GMT
• Title: Learned Hallucination Detection in Black-Box LLMs using Token-level Entropy Production Rate
• Authors: Charles Moslonka, Hicham Randrianarivo, Arthur Garnier, Emmanuel Malherbe,
• Abstract summary: Hallucinations in Large Language Model (LLM) outputs for Question Answering tasks critically undermine their real-world reliability.<n>This paper introduces an applied methodology for robust, one-shot hallucination detection, specifically designed for scenarios with limited data access.<n>Our approach derives uncertainty indicators directly from these readily available log-probabilities generated during non-greedy decoding.
• Score: 0.19676943624884313
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Hallucinations in Large Language Model (LLM) outputs for Question Answering (QA) tasks critically undermine their real-world reliability. This paper introduces an applied methodology for robust, one-shot hallucination detection, specifically designed for scenarios with limited data access, such as interacting with black-box LLM APIs that typically expose only a few top candidate log-probabilities per token. Our approach derives uncertainty indicators directly from these readily available log-probabilities generated during non-greedy decoding. We first derive an Entropy Production Rate (EPR) metric that offers baseline performance, later augmented with supervised learning. Our learned model uses features representing the entropic contributions of the accessible top-ranked tokens within a single generated sequence, requiring no multiple query re-runs. Evaluated across diverse QA datasets and multiple LLMs, this estimator significantly improves hallucination detection over using EPR alone. Crucially, high performance is demonstrated using only the typically small set of available log-probabilities (e.g., top <10 per token), confirming its practical efficiency and suitability for these API-constrained deployments. This work provides a readily deployable technique to enhance the trustworthiness of LLM responses from a single generation pass in QA and Retrieval-Augmented Generation (RAG) systems, with its utility further demonstrated in a finance framework analyzing responses to queries on annual reports from an industrial dataset.

Related papers

• Visualizing token importance for black-box language models [48.747801442240565]
  We consider the problem of auditing black-box large language models (LLMs) to ensure they behave reliably when deployed in production settings.<n>We propose Distribution-Based Sensitivity Analysis (DBSA) to evaluate the sensitivity of the output of a language model for each input token.
  arXiv  Detail & Related papers  (2025-12-12T14:01:43Z)
• Consistency Is the Key: Detecting Hallucinations in LLM Generated Text By Checking Inconsistencies About Key Facts [21.081815261690444]
  Large language models (LLMs) often hallucinate and generate text that is factually incorrect and not grounded in real-world knowledge.<n>This poses serious risks in domains like healthcare, finance, and customer support.<n>We introduce CONFACTCHECK, an efficient detection approach that does not leverage any external knowledge base.
  arXiv  Detail & Related papers  (2025-11-15T14:33:02Z)
• LaSeR: Reinforcement Learning with Last-Token Self-Rewarding [54.72617309922891]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a core paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs)<n>Previous practice requires the LLM to sequentially generate solutions and self-verifications using two separate prompt templates, which significantly reduces efficiency.<n>We propose LaSeR (Reinforcement Learning with Last-Token Self-Rewarding), an algorithm that simply augments the original RLVR loss with a MSE loss.
  arXiv  Detail & Related papers  (2025-10-16T17:55:11Z)
• IDA-Bench: Evaluating LLMs on Interactive Guided Data Analysis [60.32962597618861]
  IDA-Bench is a novel benchmark evaluating large language models in multi-round interactive scenarios.<n>Agent performance is judged by comparing its final numerical output to the human-derived baseline.<n>Even state-of-the-art coding agents (like Claude-3.7-thinking) succeed on 50% of the tasks, highlighting limitations not evident in single-turn tests.
  arXiv  Detail & Related papers  (2025-05-23T09:37:52Z)
• Unleashing the Power of LLMs in Dense Retrieval with Query Likelihood Modeling [69.84963245729826]
  We propose an auxiliary task of QL to enhance the backbone for subsequent contrastive learning of the retriever.<n>We introduce our model, which incorporates two key components: Attention Block (AB) and Document Corruption (DC)
  arXiv  Detail & Related papers  (2025-04-07T16:03:59Z)
• Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs [60.881609323604685]
  Large Language Models (LLMs) accessed via black-box APIs introduce a trust challenge.<n>Users pay for services based on advertised model capabilities.<n> providers may covertly substitute the specified model with a cheaper, lower-quality alternative to reduce operational costs.<n>This lack of transparency undermines fairness, erodes trust, and complicates reliable benchmarking.
  arXiv  Detail & Related papers  (2025-04-07T03:57:41Z)
• Learning on LLM Output Signatures for gray-box Behavior Analysis [52.81120759532526]
  Large Language Models (LLMs) have achieved widespread adoption, yet our understanding of their behavior remains limited.<n>We develop a transformer-based approach to process contamination and data detection in gray-box settings.<n>Our approach achieves superior performance on hallucination and data detection in gray-box settings, significantly outperforming existing baselines.
  arXiv  Detail & Related papers  (2025-03-18T09:04:37Z)
• Detecting Hallucinations in Large Language Model Generation: A Token Probability Approach [0.0]
  Large Language Models (LLMs) produce inaccurate outputs, also known as hallucinations. This paper introduces a supervised learning approach employing only four numerical features derived from tokens and vocabulary probabilities obtained from other evaluators. The method yields promising results, surpassing state-of-the-art outcomes in multiple tasks across three different benchmarks.
  arXiv  Detail & Related papers  (2024-05-30T03:00:47Z)
• HypoTermQA: Hypothetical Terms Dataset for Benchmarking Hallucination Tendency of LLMs [0.0]
  Hallucinations pose a significant challenge to the reliability and alignment of Large Language Models (LLMs) This paper introduces an automated scalable framework that combines benchmarking LLMs' hallucination tendencies with efficient hallucination detection. The framework is domain-agnostic, allowing the use of any language model for benchmark creation or evaluation in any domain.
  arXiv  Detail & Related papers  (2024-02-25T22:23:37Z)
• ReEval: Automatic Hallucination Evaluation for Retrieval-Augmented Large Language Models via Transferable Adversarial Attacks [91.55895047448249]
  This paper presents ReEval, an LLM-based framework using prompt chaining to perturb the original evidence for generating new test cases. We implement ReEval using ChatGPT and evaluate the resulting variants of two popular open-domain QA datasets. Our generated data is human-readable and useful to trigger hallucination in large language models.
  arXiv  Detail & Related papers  (2023-10-19T06:37:32Z)

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.