HD-NDEs: Neural Differential Equations for Hallucination Detection in LLMs
Li, Qing Geng, Jiahui Chen, Zongxiong Zhu, Derui Wang, Yuxia Ma, Congbo Lyu, Chenyang Karray, Fakhry O.
Li, Qing
Geng, Jiahui
Chen, Zongxiong
Zhu, Derui
Wang, Yuxia
Ma, Congbo
Lyu, Chenyang
Karray, Fakhry O.
Supervisor
Department
Machine Learning
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
In recent years, large language models (LLMs) have made remarkable advancements, yet hallucination, where models produce inaccurate or non-factual statements, remains a significant challenge for real-world deployment. Although current classification-based methods, such as SAPLMA, are highly efficient in mitigating hallucinations, they struggle when non-factual information arises in the early or mid-sequence of outputs, reducing their reliability. To address these issues, we propose Hallucination Detection-Neural Differential Equations (HD-NDEs), a novel method that systematically assesses the truthfulness of statements by capturing the full dynamics of LLMs within their latent space. Our approaches apply neural differential equations (Neural DEs) to model the dynamic system in the latent space of LLMs. Then, the sequence in the latent space is mapped to the classification space for truth assessment. The extensive experiments across five datasets and six widely used LLMs demonstrate the effectiveness of HD-NDEs, especially, achieving over 14% improvement in AUC-ROC on the True-False dataset compared to state-of-the-art techniques.
Citation
Q. Li et al., “HD-NDEs: Neural Differential Equations for Hallucination Detection in LLMs,” vol. 1, pp. 6173–6186, Aug. 2025, doi: 10.18653/V1/2025.ACL-LONG.309
Source
Proceedings of the Annual Meeting of the Association for Computational Linguistics
Conference
63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025
Keywords
Prompt Optimization, Large Language Models, Exemplar-Guided Reflection, Memory in Prompting, Efficient Prompt Search, Instruction Tuning, Performance Boost, Real-world NLP Tasks
Subjects
Source
63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025
Publisher
Association for Computational Linguistics