FinGEAR: Financial Mapping-Guided Enhanced Answer Retrieval
Ying Li, Mengyu Wang, Miguel de Carvalho, Sotirios Sabanis, Tiejun Ma · 2025
Read FinGEAR: Financial Mapping-Guided Enhanced Answer Retrieval by Ying Li, Mengyu Wang, Miguel de Carvalho, Sotirios Sabanis, Tiejun Ma (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-2509-12042.txt.
Description
Finding the right passage inside a long company filing is hard. Ying Li and 4 co-authors propose FinGEAR in a 2025 paper to improve retrieval, the search step before a language model answers a question. Filings such as 10-K reports are long, follow a strict section hierarchy, and use specialized financial language.
FinGEAR uses a finance word list to guide search within each numbered section, builds two kinds of hierarchical indexes, and then reranks results in two stages. Tested on full 10-K filings with questions from the FinQA dataset, it raised precision, recall, and related scores over flat, graph-based, and tree-based search. Answer accuracy also rose with the same reader model. The design offers a practical base for high-stakes financial analysis.
Catalog record
| CREATOR | Ying Li, Mengyu Wang, Miguel de Carvalho, Sotirios Sabanis, Tiejun Ma |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-12042