FacLens: Transferable Probe for Foreseeing Non-Factuality in Fact-Seeking Question Answering of Large Language Models

TL;DR

This paper introduces FacLens, a lightweight and transferable model that predicts non-factual responses in large language models before they generate answers, improving efficiency and transferability over previous methods.

Contribution

The work presents a novel, efficient probe for non-factuality prediction that transfers across different LLMs, reducing development costs and enhancing prediction accuracy.

Findings

FacLens outperforms existing methods in effectiveness and efficiency.

Hidden question representations show similar NFP patterns across LLMs.

FacLens enables transferability of non-factuality prediction across models.

Abstract

Despite advancements in large language models (LLMs), non-factual responses still persist in fact-seeking question answering. Unlike extensive studies on post-hoc detection of these responses, this work studies non-factuality prediction (NFP), predicting whether an LLM will generate a non-factual response prior to the response generation. Previous NFP methods have shown LLMs' awareness of their knowledge, but they face challenges in terms of efficiency and transferability. In this work, we propose a lightweight model named Factuality Lens (FacLens), which effectively probes hidden representations of fact-seeking questions for the NFP task. Moreover, we discover that hidden question representations sourced from different LLMs exhibit similar NFP patterns, enabling the transferability of FacLens across different LLMs to reduce development costs. Extensive experiments highlight FacLens's superiority in both effectiveness and efficiency.