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.
MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs
Tsinghua University(清华大学);Tsinghua University, Tencent(清华大学腾讯);Tencent(腾讯);University of Science and Technology Beijing(北京科技大学);University of Macau(澳门大学)
Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction
Ministry of Education Key Laboratory of Intelligent Networks and Network Security(教育部智能网络与网络安全重点实验室);Centre for Frontier AI Research, Institute of High Performance Computing, Agency for Science, Technology and Research(前沿人工智能研究中心,高性能计算研究所,科技研究局)
Mitigating Object Hallucinations in Large Vision-Language Models with Assembly of Global and Local Attention
Xi’an Jiaotong University(西安交通大学);National Engineering Laboratory for Big Data Analytics(国家大数据分析工程实验室);Nanyang Technological University(南洋理工大学);Lenovo Research(联想研究院);University of Massachusetts Boston(马萨诸塞大学波士顿分校);Zhejiang University of Technology(浙江工业大学)
RAL2M: Retrieval Augmented Learning-To-Match Against Hallucination in Compliance-Guaranteed Service Systems
Hong Kong Polytechnic University(香港理工大学);New York University Shanghai(纽约大学上海分校);Hong Kong University of Science and Technology(香港科技大学);Beijing University of Posts and Telecommunications(北京邮电大学)
Align Generative Artificial Intelligence with Human Preferences: A Novel Large Language Model Fine-Tuning Method for Online Review Management
Department of Information Systems and Operations Management, The University of Texas at Arlington(信息系统与运营管理系,德克萨斯大学阿灵顿分校);Department of Management Information Systems, University of Arizona(管理信息学系,亚利桑那大学)
Learning Robust Representations Of Generative Models Using Set-Based Artificial Fingerprints
USC Information Sciences Institute(南加州大学信息科学研究所);Visual Intelligence and Multimedia Analytics Laboratory(视觉智能与多媒体分析实验室);USC Ming Hsieh Department of Electrical and Computer Engineering(南加州大学明尼苏达电子与计算机工程系)
AI and Authenticity in Islamic Research: A Critical Evaluation of Generative AI Reliability, Hallucination, and Source Fidelity in Quranic, Hadith, and Fiqh Knowledge
School of Computer Science, The University of Sydney(悉尼大学计算机学院);Macquarie University(麦考瑞大学);The University of New South Wales(新南威尔士大学)
The Creative Frontier of Generative AI: Managing the Novelty-Usefulness Tradeoff
Samuel Curtis Johnson Graduate School of Management, Cornell University(康奈尔大学塞缪尔·柯蒂斯·约翰逊管理研究生院);Lee Kong Chian School of Business, Singapore Management University(新加坡管理大学李光前商学院)
GIRL: Generative Imagination Reinforcement Learning via Information-Theoretic Hallucination Control
Department of Computer Science and Engineering, Visvesvaraya Technological University (VTU), Belagavi, India(维斯瓦拉亚科技大学计算机科学与工程系,贝拉加维,印度);Aliens on Earth (AoE) Autonomous Research Group, Belagavi, India(地球外星人自主研究组,贝拉加维,印度)
Latent fingerprint enhancement for accurate minutiae detection
TCC Research and Development Labs, Technology Control Company(TCC研发实验室,技术控制公司);University of New South Wales(新南威尔士大学);Abasyn University Islamabad(阿巴辛伊斯兰堡大学);Computer and Information Sciences College, Jouf University(焦夫大学计算机与信息科学学院)
Leveraging generative hallucination and biophysics-informed modeling for unified biomolecular sequence-structure co-design
University of Florida(佛罗里达大学);University of Chicago(芝加哥大学);Data Science Institute, University of Chicago(芝加哥大学数据科学研究所);Toyota Technological Institute at Chicago(芝加哥丰田技术研究所);Mayo Clinic(梅奥诊所);Argonne National Laboratory(阿贡国家实验室)
Center for Foundation Models and Generative AI, Northwestern University(基础模型与生成式人工智能中心,西北大学);Department of Computer Science, Northwestern University(计算机科学系,西北大学);Ensemble AI;Engineering Science, University of Toronto(工程科学,多伦多大学);University of California, Berkeley(加州大学伯克利分校);Department of Statistics and Data Science, Northwestern University(统计与数据科学系,西北大学)