UNCLE: Benchmarking Uncertainty Expressions in Long-Form Generation
This addresses the hallucination problem in LLMs for users needing reliable long-form generation, though it's incremental as it focuses on evaluation rather than solving hallucination directly.
The authors tackled the problem of evaluating LLMs' ability to express uncertainty in long-form generation by introducing UNCLE, a benchmark with over 1,000 entities across five domains, and found that current models fail to convey uncertainty appropriately in this context.
Large Language Models (LLMs) are prone to hallucination, particularly in long-form generations. A promising direction to mitigate hallucination is to teach LLMs to express uncertainty explicitly when they lack sufficient knowledge. However, existing work lacks direct and fair evaluation of LLMs' ability to express uncertainty effectively in long-form generation. To address this gap, we first introduce UNCLE, a benchmark designed to evaluate uncertainty expression in both long- and short-form question answering (QA). UNCLE covers five domains and includes more than 1,000 entities, each with paired short- and long-form QA items. Our dataset is the first to directly link short- and long-form QA through aligned questions and gold-standard answers. Along with UNCLE, we propose a suite of new metrics to assess the models' capabilities to selectively express uncertainty. We then demonstrate that current models fail to convey uncertainty appropriately in long-form generation. We further explore both prompt-based and training-based methods to improve models' performance, with the training-based methods yielding greater gains. Further analysis of alignment gaps between short- and long-form uncertainty expression highlights promising directions for future research using UNCLE.