HiFi-KPI: A Dataset for Hierarchical KPI Extraction from Earnings Filings

arXiv cs.CL / 3/20/2026

📰 NewsTools & Practical UsageModels & Research

共有:

Key Points

Introduces the HiFi-KPI dataset, a large-scale resource for hierarchical KPI extraction from earnings filings, comprising 1.65M paragraphs and 198k hierarchical labels linked to iXBRL taxonomies.
Defines three evaluation tasks (KPI classification, KPI extraction, and structured KPI extraction) and releases HiFi-KPI-Lite, a manually curated 8K-paragraph subset.
Reports strong baselines: encoder-based models reach over 0.906 macro-F1 on classification, while LLMs achieve about 0.440 F1 on structured extraction, with most errors tied to date handling.
Open-sources all code and data at the provided GitHub repository, facilitating reproducibility and further research.
Aims to improve cross-company transferability of KPI tagging in financial filings and accelerate rapid evaluation for KPI extraction systems.

Abstract

Accurate tagging of earnings reports can yield significant short-term returns for stakeholders. The machine-readable inline eXtensible Business Reporting Language (iXBRL) is mandated for public financial filings. Yet, its complex, fine-grained taxonomy limits the cross-company transferability of tagged Key Performance Indicators (KPIs). To address this, we introduce the Hierarchical Financial Key Performance Indicator (HiFi-KPI) dataset, a large-scale corpus of 1.65M paragraphs and 198k unique, hierarchically organized labels linked to iXBRL taxonomies. HiFi-KPI supports multiple tasks and we evaluate three: KPI classification, KPI extraction, and structured KPI extraction. For rapid evaluation, we also release HiFi-KPI-Lite, a manually curated 8K paragraph subset. Baselines on HiFi-KPI-Lite show that encoder-based models achieve over 0.906 macro-F1 on classification, while Large Language Models (LLMs) reach 0.440 F1 on structured extraction. Finally, a qualitative analysis reveals that extraction errors primarily relate to dates. We open-source all code and data at https://github.com/aaunlp/HiFi-KPI.