SEMANTIC REFORMULATION ENTROPY FOR ROBUST HALLUCINATION DETECTION IN QA TASKS
Chaodong Tong, School of Cyber Security, Zhang Qi, Lei Jiang, Yanbing Liu, School of Cyber Security, Nannan Sun, Wei Li
Abstract
Reliable question answering with large language models (LLMs) is challenged by hallucinations, fluent but factually incorrect outputs arising from epistemic uncertainty. Existing entropy-based semantic-level uncertainty estimation methods are limited by sampling noise and unstable clustering of variable-length answers. We propose Semantic Reformulation Entropy (SRE), which improves uncertainty estimation in two ways. First, input-side semantic reformulations produce faithful paraphrases, expand the estimation space, and reduce biases from superficial decoder tendencies. Second, progressive, energy-based hybrid clustering stabilizes semantic grouping. Experiments on SQuAD and TriviaQA show that SRE outperforms strong baselines, providing more robust and generalizable hallucination detection. These results demonstrate that combining input diversification with multi-signal clustering substantially enhances semantic-level uncertainty estimation.
BibTeX
@inproceedings{icassp2026_semanticreformul,
title = {SEMANTIC REFORMULATION ENTROPY FOR ROBUST HALLUCINATION DETECTION IN QA TASKS},
author = {Chaodong Tong and School of Cyber Security and Zhang Qi and Lei Jiang and Yanbing Liu and School of Cyber Security and Nannan Sun and Wei Li},
booktitle = {ICASSP 2026},
year = {2026}
}