Reproduce “Effective Performance Measurement: Challenges and Opportunities in KPI Extraction from Earnings Calls” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Rasmus T. Aavang, Rasmus Tjalk-Bøggild, Alexandre Iolov, Giovanni Rizzi, Mike Zhang, Johannes Bjerva
The authors develop a method to extract and track key performance indicators (KPIs) in earnings calls without a predefined taxonomy of KPI names. It prompts language models to identify metric descriptions and values, then links and groups them across calls. On the small annotated ECB-A benchmark, Gemini 3 Pro reaches 11.5% exact F1, versus 0.0% for evaluated SEC-trained encoders.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.