Improving the Completeness and Comparability of Segment Disclosures: A Large Language Model Approach

A new large language model-based framework aims to extract segment disclosures directly from Form 10-K filings, addressing long-standing completeness and comparability challenges that have limited empirical financial research relying on structured databases [1]. Segment-level disclosures are a central component of financial reporting, offering insight into how firms organize internally and allocate economic activities across operating units [1]. The information is often presented in both qualitative and quantitative forms, dispersed across tables and narrative sections of Form 10-K filings [1]. Empirical research that depends on structured databases has faced persistent completeness and comparability challenges: some firm-year observations may be missing, nested segment disclosures are not captured, and support for longitudinal and cross-firm comparability is limited [1]. The study, submitted on 20 April 2026, develops a large language model-based framework to extract segment disclosures directly from Form 10-K filings and to preserve both reportable and nested segment information [1]. The researchers further designed a retrieval augmented system that incorporates information across multiple filings to support comparability [1]. Two representative settings demonstrate the framework’s application: longitudinal analysis within a firm to interpret segment changes over time, and cross-firm alignment of geographic segments across firms with different reporting structures [1]. The results indicate that the artifact accurately extracts segment-level information and effectively addresses questions that require cross-period knowledge [1]. The work highlights the potential of LLM-based approaches to enhance the measurement and interpretation of segment disclosures, a domain where traditional structured databases have fallen short [1]. The framework was developed under arXivLabs, a platform that allows collaborators to develop and share new arXiv features directly on the website, with commitments to openness, community, excellence, and user data privacy [1].