可扩展的令牌级幻觉检测方法在大语言模型中
Scalable Token-Level Hallucination Detection in Large Language Models
AI总结：
本文提出TokenHD方法，通过可扩展的数据引擎和重要性加权策略，实现令牌级幻觉检测，实验显示小型模型在训练后性能优于大模型，且检测能力随模型规模扩大而提升。
摘要：
大型语言模型（LLMs）展现出卓越的能力，但仍然频繁产生幻觉。这些幻觉在推理密集型任务中难以检测，内容看似连贯但包含逻辑错误和不可靠的中间结果。尽管步级分析常用于检测内部幻觉，但其受限于粒度有限和可扩展性差，因为依赖于步分割。为解决这些限制，我们提出TokenHD，一个全面的训练管道用于训练令牌级幻觉检测器。具体而言，TokenHD包含一个可扩展的数据引擎用于合成大规模幻觉注释，以及一个训练配方，其特征是重要性加权策略用于稳健的模型训练。为了系统评估检测性能，我们还提供了一个严谨的评估协议。通过在TokenHD中训练，我们的检测器直接在自由文本上识别幻觉，消除了预定义步分割或额外文本重新格式化的需要。我们的实验显示，即使是一个小型检测器（0.6B）在训练后也实现了显著的性能提升，超越了许多较大的推理模型（例如QwQ-32B），并且检测性能从0.6B到8B的模型规模上保持一致的扩展性。最后，我们展示我们的检测器能够很好地泛化到多样的实际场景，并探索进一步增强其跨领域泛化能力的策略。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
