Step-by-Step Fact Verification System for Medical Claims with Explainable Reasoning
This work addresses the problem of verifying domain-specific medical claims for healthcare professionals and researchers, though it is incremental as it extends existing iterative methods to a new domain.
The researchers tackled fact verification for medical claims by applying an iterative, step-by-step system using large language models, which improved performance over traditional methods on three medical datasets.
Fact verification (FV) aims to assess the veracity of a claim based on relevant evidence. The traditional approach for automated FV includes a three-part pipeline relying on short evidence snippets and encoder-only inference models. More recent approaches leverage the multi-turn nature of LLMs to address FV as a step-by-step problem where questions inquiring additional context are generated and answered until there is enough information to make a decision. This iterative method makes the verification process rational and explainable. While these methods have been tested for encyclopedic claims, exploration on domain-specific and realistic claims is missing. In this work, we apply an iterative FV system on three medical fact-checking datasets and evaluate it with multiple settings, including different LLMs, external web search, and structured reasoning using logic predicates. We demonstrate improvements in the final performance over traditional approaches and the high potential of step-by-step FV systems for domain-specific claims.