Georgia Institute of Technology
Language Modeling in Finance: Evaluation, Extraction, and Economic Insight
Abstract
dc:description.abstractOver the past decade, advances in self-attention mechanisms and foundation models have transformed Natural Language Processing (NLP), creating new opportunities and challenges for high-stakes domains such as finance. This work positions finance as both a critical application area and a rigorous diagnostic lens for evaluating the reasoning, bias, and reliability of Large Language Models (LLMs). It uncovers structural limitations of generalist models in financial contexts and introduces calibration and weak-supervision frameworks to address these challenges. At the same time, it develops specialized domain-adapted architectures, datasets, and benchmarks for financial information extraction, institutional discourse analysis, and multimodal economic forecasting. Together, these contributions demonstrate that adapting NLP systems to the financial language enables more reliable AI systems while generating actionable economic insight with impact beyond finance.
Degree
thesis:*- Name thesis:degree_name
- Machine Learning, PhD
- Department thesis:degree_department
- Computational Science and Engineering
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Shah, Agam Alkeshkumar
- Advisors dc:contributor.advisor
-
- Chava, Sudheer
- Zhang, Chao
- Committee members dc:contributor.committeemember
-
- Prakash, B. Aditya
- Aluru, Srinivas
- Mitra, Indrajit
- ORCID person:identifier.orcid
Rights
- Language dc:language.iso
- English
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81679