Cross Paraphrastic Invariance Learning

Cross Paraphrastic Invariance Learning for Hallucination Detection

arXiv:2606.08157v1 Announce Type: new Abstract: Large language models (LLMs) frequently generate hallucinations, which are unsupported by a source document. To avoid costly LLM-as-evaluator pipelines and the heavy annotation demands of existing classifiers, we propose CPIL (Cross Paraphrastic Invariance Learning), a two-stage Siamese framework that maximizes the utility of existing labeled data. Concretely, CPIL constructs informative training pairs by: (i) generating paraphrastic views of...