Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.03147 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914528630931456 |
|---|---|
| author | Aavang, Rasmus T. Tjalk-Bøggild, Rasmus Iolov, Alexandre Rizzi, Giovanni Zhang, Mike Bjerva, Johannes |
| author_facet | Aavang, Rasmus T. Tjalk-Bøggild, Rasmus Iolov, Alexandre Rizzi, Giovanni Zhang, Mike Bjerva, Johannes |
| contents | Earnings calls are a key source of financial information about public companies. However, extracting information from these calls is difficult. Unlike the templatic filings required by the U.S. Securities and Exchange Commission (SEC) to report a company's financial situation, earnings conference calls have no built-in labels, are unstructured, and feature conversational language. We explore this challenging domain by assessing the information captured by models trained on SEC filings and in-context learning methods. To establish a baseline, we first evaluate the generalization capabilities of SEC-trained models across established SEC datasets. To support our investigation, we introduce three novel benchmarks: (1) SEC Filings Benchmark (SECB), (2) Earnings Calls Benchmark (ECB), and ECB-A, a subset with 2,460 expert annotation groups to support our qualitative analysis. We find that encoder-based models struggle with the domain shift. Finally, we propose a system utilizing LLMs to perform open-ended extraction from unstructured call transcripts, verified by human evaluation (79.7% precision), providing a baseline for this valuable domain through the consistent tracking of emergent KPIs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_03147 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Effective Performance Measurement: Challenges and Opportunities in KPI Extraction from Earnings Calls Aavang, Rasmus T. Tjalk-Bøggild, Rasmus Iolov, Alexandre Rizzi, Giovanni Zhang, Mike Bjerva, Johannes Computation and Language Earnings calls are a key source of financial information about public companies. However, extracting information from these calls is difficult. Unlike the templatic filings required by the U.S. Securities and Exchange Commission (SEC) to report a company's financial situation, earnings conference calls have no built-in labels, are unstructured, and feature conversational language. We explore this challenging domain by assessing the information captured by models trained on SEC filings and in-context learning methods. To establish a baseline, we first evaluate the generalization capabilities of SEC-trained models across established SEC datasets. To support our investigation, we introduce three novel benchmarks: (1) SEC Filings Benchmark (SECB), (2) Earnings Calls Benchmark (ECB), and ECB-A, a subset with 2,460 expert annotation groups to support our qualitative analysis. We find that encoder-based models struggle with the domain shift. Finally, we propose a system utilizing LLMs to perform open-ended extraction from unstructured call transcripts, verified by human evaluation (79.7% precision), providing a baseline for this valuable domain through the consistent tracking of emergent KPIs. |
| title | Effective Performance Measurement: Challenges and Opportunities in KPI Extraction from Earnings Calls |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.03147 |