IUQ: Interrogative Uncertainty Quantification for Long-Form Large Language Model Generation

TL;DR

This paper introduces IUQ, a new framework for quantifying uncertainty in long-form LLM outputs, addressing factual accuracy and semantic complexity challenges.

Contribution

IUQ leverages inter-sample consistency and intra-sample faithfulness to provide reliable claim-level uncertainty measures in long-form LLM generation.

Findings

IUQ outperforms existing methods on long-form generation datasets.

The framework provides reliable claim-level uncertainty and faithfulness measures.

Experimental results show IUQ's effectiveness across diverse models.

Abstract

Despite the rapid advancement of Large Language Models (LLMs), uncertainty quantification in LLM generation is a persistent challenge. Although recent approaches have achieved strong performance by restricting LLMs to produce short or constrained answer sets, many real-world applications require long-form and free-form text generation. A key difficulty in this setting is that LLMs often produce responses that are semantically coherent yet factually inaccurate, while the underlying semantics are multifaceted and the linguistic structure is complex. To tackle this challenge, this paper introduces Interrogative Uncertainty Quantification (IUQ), a novel framework that leverages inter-sample consistency and intra-sample faithfulness to quantify the uncertainty in long-form LLM outputs. By utilizing an interrogate-then-respond paradigm, our method provides reliable measures of claim-level uncertainty and the model's faithfulness. Experimental results across diverse model families and model sizes demonstrate the superior performance of IUQ over two widely used long-form generation datasets. The code is available at https://github.com/louisfanhz/IUQ.