Fine-Tuning Large Language Models to Appropriately Abstain with Semantic Entropy
This addresses hallucination risks in critical domains like medicine or law, but is incremental as it builds on existing fine-tuning approaches.
The paper tackles the problem of LLM hallucination by fine-tuning models to abstain from answering uncertain questions, using semantic entropy as an uncertainty measure without external labels, and shows it matches or outperforms prior methods on various datasets.
Large Language Models (LLMs) are known to hallucinate, whereby they generate plausible but inaccurate text. This phenomenon poses significant risks in critical applications, such as medicine or law, necessitating robust hallucination mitigation strategies. While recent works have proposed fine-tuning methods to teach LLMs to abstain from answering questions beyond their knowledge or capabilities, these methods rely on the existence of ground-truth labels or are limited to short-form responses. To address these limitations, we propose fine-tuning using semantic entropy, an uncertainty measure derived from introspection into the model which does not require external labels. We demonstrate that our approach matches or outperforms models fine-tuned using prior work and achieves strong performance for both short and long-form generations on a range of datasets.