FDARxBench: Benchmarking Regulatory and Clinical Reasoning on FDA Generic Drug Assessment

TL;DR

FDARxBench is a comprehensive benchmark for evaluating language models' ability to understand and reason over FDA drug labels, highlighting gaps in factual accuracy, retrieval, and safe refusal.

Contribution

The paper introduces FDARxBench, a novel expert-curated benchmark with a multi-stage QA pipeline for regulatory and clinical reasoning on FDA drug labels.

Findings

Models show significant gaps in factual grounding.

Long-context retrieval remains challenging.

Safe refusal behavior is often inadequate.

Abstract

We introduce an expert curated, real-world benchmark for evaluating document-grounded question-answering (QA) motivated by generic drug assessment, using the U.S. Food and Drug Administration (FDA) drug label documents. Drug labels contain rich but heterogeneous clinical and regulatory information, making accurate question answering difficult for current language models. In collaboration with FDA regulatory assessors, we introduce FDARxBench, and construct a multi-stage pipeline for generating high-quality, expert curated, QA examples spanning factual, multi-hop, and refusal tasks, and design evaluation protocols to assess both open-book and closed-book reasoning. Experiments across proprietary and open-weight models reveal substantial gaps in factual grounding, long-context retrieval, and safe refusal behavior. While motivated by FDA generic drug assessment needs, this benchmark also provides a substantial foundation for challenging regulatory-grade evaluation of label comprehension. The benchmark is designed to support evaluation of LLM behavior on drug-label questions.