Agentic Retrieval of Topics and Insights from Earnings Calls Conference: FinIR: The 2nd Workshop on Financial Information Retrieval in the Era of Generative AI; July 17, 2025; Padua, Italy
Anant Gupta, Rajarshi Bhowmik, Geoffrey Gunow · 2025
Read Agentic Retrieval of Topics and Insights from Earnings Calls Conference: FinIR: The 2nd Workshop on Financial Information Retrieval in the Era of Generative AI; July 17, 2025; Padua, Italy by Anant Gupta, Rajarshi Bhowmik, Geoffrey Gunow (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-07906.txt.
Description
Anant Gupta, Rajarshi Bhowmik, and Geoffrey Gunow use language model agents to track what companies talk about in their earnings calls. Financial analysts follow strategic themes over time, but traditional topic models struggle as industries evolve and new topics emerge. Their agent extracts topics from call transcripts and organizes them into a hierarchical ontology.
The system links new topics to existing ones, so the ontology grows with the calls each quarter. The extracted topics then support company-level insights and trend detection over time. Evaluation measures ontology coherence, how accurately topic evolution is captured, and the ability to surface emerging financial trends. The authors present the agent as a dynamic alternative to static topic modeling for finance.
Catalog record
| CREATOR | Anant Gupta, Rajarshi Bhowmik, Geoffrey Gunow |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-07906