Agent AI for Finance
Description
This open access book provides an overview of the current state of financial argument mining and financial text generation, and presents the authors' thoughts on the blueprint for NLP in finance in the agent AI era.
Financial documents contain numerous causal inferences and subjective opinions. In a previous book, "From Opinion Mining to Financial Argument Mining" (Springer, 2021), the first author discussed understanding financial documents in a fine-grained manner, particularly those containing opinions. The book highlighted several future directions, such as financial argument mining, multimodal opinion understanding, and analysis generation, and anticipated a lengthy journey for these topics. However, since 2022, ChatGPT and large language models (LLMs) have shown promising advancements, motivating the authors to write this second book on the topic of financial Natural Language Processing (NLP).
Agent-based AI systems have been widely discussed since the advent of LLMs. This book aims to equip researchers and practitioners with the latest methodologies, concepts, and frameworks for developing, deploying, and evaluating AI agents with capabilities in multimodal understanding, decision-making, and interaction. It places a special emphasis on human-centered decision-making and multi-agent cooperation in financial applications. The book surveys the current landscape and discuss future research and development directions.
Targeting a wide audience, from students to seasoned researchers in AI and finance, this book offers an overview of recent trends in Agent AI for finance. It provides a foundation for students to understand the field and design their research direction, while inviting experienced researchers to engage in discussions on open research questions informed by pilot experimental results.
Although this book focuses on financial applications, the discussed concepts and methods can also be applied to other real-world applications by integrating domain-specific characteristics. The authors look forward to seeing new findings and more novel extensions based on the proposed ideas.
More details
Persons
Chung-Chi Chen is currently a researcher at the Artificial Intelligence Research Center, AIST, Japan. His scholarly pursuits revolve around the intricate realm of financial opinion mining and the nuanced understanding and generation of financial documents. He is the founder of ACL SIG-FinTech, and he has orchestrated the FinNLP/FinWeb workshop series within prestigious conferences such as IJCAI, WWW, EMNLP, and IJCNLP-AACL since 2019. He has guided the FinNum and FinArg shared task series on the NTCIR since 2018. He was also a presenter in the AACL-2020, EMNLP-2021, ECAI-2024, and SIGIR-2025 tutorials. He served as a program co-chair of NTCIR-18, senior area chair of ACL-2024, and PC member in many representative conferences. In academic competitions, he won the SIGIR Early Career Researcher Award (Excellence in Community Engagement), in addition to two Thesis Awards and Technology Innovation Award. Outside academia, he has actively explored the fast-paced FinTech industry, winning multiple awards in startup, FinTech, and LegalTech competitions.
Hiroya Takamura is the research team leader at Knowledge and Information Research Team at the Artificial Intelligence Research Center, AIST, Japan. His research interest includes sentiment analysis, text summarization, and natural language generation. He has authored a number of papers regarding with knowledge extraction for numerical attributes, and understanding and generating numbers in text. In addition, he has experience in organizing conferences and workshops. He was a member of the organization committee of AACL. He served as a program chair of International Conference on Natural Language Generation (INLG) 2019. He has also organized several workshops. He served as an area chair of some conferences (EMNLP, EACL, COLING) and as a PC member of many conferences (ACL, NAACL, EMNLP, EACL, COLING, and so on).
Content
Preface.- 1. Introduction.- 2. Financial Argument Mining.- 3. Single-Agent/Model Design.- 4. Multi-Agent Interaction.- 5. Multi-Scale Model Synergy.- 6. Generative AI Application Scenarios.- 7. Looking to the Future.