Scalable Token-Level Hallucination Detection in Large Language Models

2026-05-12 · Rui Min, Tianyu Pang, Chao Du, Minhao Cheng, Yi R. Fung arxiv

Large language models (LLMs) have demonstrated remarkable capabilities, but they still frequently produce hallucinations. These hallucinations are difficult to detect in reasoning-intensive tasks, where the content appears coherent but contains errors like logical flaws and unreliable intermediate results. While step-level analysis is commonly used to detect internal hallucinations, it suffers from limited granularity and poor scalability due to its reliance on step segmentation. To address these limitations, we propose TokenHD, a holistic pipeline for training token-level hallucination detectors. Specifically, TokenHD consists of a scalable data engine for synthesizing large-scale hallucination annotations along with a training recipe featuring an importance-weighted strategy for robust model training. To systematically assess the detection performance, we also provide a rigorous evaluation protocol. Through training within TokenHD, our detector operates directly on free-form text to identify hallucinations, eliminating the need for predefined step segmentation or additional text reformatting. Our experiments show that even a small detector (0.6B) achieves substantial performance gains after training, surpassing much larger reasoning models (e.g., QwQ-32B), and detection performance scales consistently with model size from 0.6B to 8B. Finally, we show that our detector can generalize well across diverse practical scenarios and explore strategies to further enhance its cross-domain generalization capability.

📄 PDF Abstract BibTeX arXiv:2605.12384

Code (0)

등록된 구현이 없습니다.

Tasks

Domain Generalization

Real-Time Detection of Hallucinated Entities in Long-Form Generation

2025-08-26 · Oscar Obeso, Andy Arditi, Javier Ferrando, Joshua Freeman 외 arxiv

Large language models are now routinely used in high-stakes applications where hallucinations can cause serious harm, such as medical consultations or legal advice. Existing hallucination detection methods, however, are …

Mathematical Reasoning

First Hallucination Tokens Are Different from Conditional Ones

2025-07-28 · Jakob Snel, Seong Joon Oh arxiv

Large Language Models (LLMs) hallucinate, and detecting these cases is key to ensuring trust. While many approaches address hallucination detection at the response or span level, recent work explores token-level detectio…

Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding

2026-01-29 · Yifan Zhu, Huiqiang Rong, Haoran Luo arxiv

Large Language Models (LLMs) often hallucinate, generating content inconsistent with the input. Retrieval-Augmented Generation (RAG) and Reinforcement Learning with Human Feedback (RLHF) can mitigate hallucinations but r…

Reinforcement Learning

Measuring the Impact of Lexical Training Data Coverage on Hallucination Detection in Large Language Models

2025-11-22 · Shuo Zhang, Fabrizio Gotti, Fengran Mo, Jian-Yun Nie arxiv

Hallucination in large language models (LLMs) is a fundamental challenge, particularly in open-domain question answering. Prior work attempts to detect hallucination with model-internal signals such as token-level entrop…

Open-Domain Question Answering

ANHALTEN: Cross-Lingual Transfer for German Token-Level Reference-Free Hallucination Detection

2024-07-18 · Janek Herrlein, Chia-Chien Hung, Goran Glavaš

Research on token-level reference-free hallucination detection has predominantly focused on English, primarily due to the scarcity of robust datasets in other languages. This has hindered systematic investigations into t…

Cross-Lingual Transfer Hallucination