AI for Finance Symposium ’25
The 2nd Workshop on
LLMs and Generative AI for Finance
AI for Finance Symposium ’25
The 2nd Workshop on LLMs and Generative AI for Finance
AI for Finance Symposium ’25
The 2nd Workshop on
LLMs and Generative AI for Finance
Workshop at ACM ICAIF '25
Workshop at ACM ICAIF '25
Submit Your Paper
Key Dates
Submission Deadline: October 1st, 2025 (AoE)
Author Notification: October 15th, 2025 (AoE)
Workshop: November 15th, 2025 (SGT)
Workshop: Morning, Nov 15th (Sat), 2025
Submission Deadline: Oct 1st, 2025 (AoE)
Author Notification: Oct 15th, 2025 (AoE)
Workshop: Nov 15th, 2025 (SGT)
Location
Singapore – Venue to be announced
Singapore – Venue to be announced
Workshop Contact
Dr. Jacob Chanyeol Choi
Dr. Jacob Chanyeol Choi
contact@ai4f.org
contact@ai4f.org
Sponsored by
Sponsored by
Speakers
Speakers
James Sullivan
James Sullivan
JPMorgan Chase
JPMorgan Chase
Venugopal Garre
Venugopal Garre
AB Bernstein
AB Bernstein
Shaun Cochran
Shaun Cochran
CLSA
CLSA
Prof. Sy Bor Wang
Prof. Sy Bor Wang
National University of Singapore
National University of Singapore
Prof. Antoine Didisheim
Prof. Antoine Didisheim
University of Melbourne
University of Melbourne
Dr. Igor Halperin
Dr. Igor Halperin
Fidelity Investments
Fidelity Investments
Sandy Chen Dowling
Sandy Chen Dowling
CLSA
CLSA
Organizing Committee
Organizing Committee
Dr. Jacob Chanyeol Choi
Dr. Jacob Chanyeol Choi
LinqAlpha
United States
Prof. Yoon Kim
Prof. Yoon Kim
MIT
United States
Prof. Alejandro
Lopez Lira
Prof. Alejandro
Lopez Lira
University of Florida
United States
Prof. Atlas Wang
Prof. Atlas Wang
UT Austin & XTX Markets
United States
Dr. Dhagash Mehta
Dr. Dhagash Mehta
BlackRock
United States
Dr. Igor Halperin
Dr. Igor Halperin
Fidelity Investments
United States
Andrew Chin
Andrew Chin
AllianceBernstein
United States
Dr. Edward Tong
Dr. Edward Tong
Google
United States
Prof. Yongjae Lee
Prof. Yongjae Lee
UNIST
South Korea
Joo Lee
Joo Lee
Arrowpoint Investment Partners
Singapore
Arrowpoint Investment
Partners, Singapore
Prof. Sy Bor Wang
Prof. Sy Bor Wang
National University of Singapore Singapore
National University of Singapore, Singapore
Dr. V. Zach Golkhou
Dr. V. Zach Golkhou
J.P. Morgan Chase
United States
Prof. Chee Seng Chan
Prof. Chee Seng Chan
Universiti Malaya
Malaysia
Dr. Lixin Fan
Dr. Lixin Fan
WeBank
China
Dr. Georgios Papaioannou
Dr. Georgios Papaioannou
Qube Research & Technologies
UK
About
the workshop
About the workshop
The rapid advancements in large language models (LLMs) and generative AI more broadly have significantly impacted the interpretation and utilization of unstructured financial data, enabling a wide range of applications in finance and accounting. While LLMs have unlocked new possibilities in this domain, NLP techniques more broadly remain vital for various applications such as financial analysis, case studies, price forecasting, portfolio optimization, and analyzing financial reports and alternative data. They are also crucial in financial risk modeling, such as credit assessment, bankruptcy detection, and M&A target prediction using news sentiment and topic detection.
However, LLMs and NLP techniques face challenges in the financial domain, including the limited context window of LLMs, the complexities of converting unstructured data into structured formats, hallucinations, and the need for interpretability in financial applications. This workshop aims to bridge the gap between technological advancements in computer science and the specific needs of finance and accounting. By focusing on methodologies such as fine-tuning, retrieval-augmented generation (RAG), prompt engineering, and NLP, we aim to showcase how these techniques can improve the accuracy and relevance of insights derived from financial and accounting data. The ultimate goal is to foster interdisciplinary collaboration, enhance industry solutions, and contribute to academic research in finance and accounting.
The workshop will be held in conjunction with ICAIF ’25.
The rapid advancements in large language models (LLMs) and generative AI more broadly have significantly impacted the interpretation and utilization of unstructured financial data, enabling a wide range of applications in finance and accounting. While LLMs have unlocked new possibilities in this domain, NLP techniques more broadly remain vital for various applications such as financial analysis, case studies, price forecasting, portfolio optimization, and analyzing financial reports and alternative data. They are also crucial in financial risk modeling, such as credit assessment, bankruptcy detection, and M&A target prediction using news sentiment and topic detection.
However, LLMs and NLP techniques face challenges in the financial domain, including the limited context window of LLMs, the complexities of converting unstructured data into structured formats, hallucinations, and the need for interpretability in financial applications. This workshop aims to bridge the gap between technological advancements in computer science and the specific needs of finance and accounting. By focusing on methodologies such as fine-tuning, retrieval-augmented generation (RAG), prompt engineering, and NLP, we aim to showcase how these techniques can improve the accuracy and relevance of insights derived from financial and accounting data. The ultimate goal is to foster interdisciplinary collaboration, enhance industry solutions, and contribute to academic research in finance and accounting.
The workshop will be held in conjunction with ICAIF ’25.
There is no length limit or strict format requirement. While formatting according to ACM guidelines is recommended (see below), we encourage submissions in an open format to allow the authors to focus on content with formatting restrictions. Submissions should be in PDF format and can be directly submitted as they are.
The workshop is non-archival and will not have official proceedings.
Call for Papers
Submit
We welcome papers exploring the use of Generative AI and Large Language Models (LLMs) across both quantitative and discretionary fund contexts.
Topics include the following but are not limited to:
Multi Agentic Systems
Multi Agentic Systems
Exploring how multi-agent systems can collaborate to streamline decision-making, analysis, and research across diverse financial tasks.
Exploring how multi-agent systems can collaborate to streamline decision-making, analysis, and research across diverse financial tasks.
Composable and Compound System Design for Explainable Reasoning
Composable and Compound System Design for Explainable Reasoning
Developing modular architectures that integrate multiple reasoning components to enhance transparency, adaptability, and interpretability in complex decision-making.
Developing modular architectures that integrate multiple reasoning components to enhance transparency, adaptability, and interpretability in complex decision-making.
Trust and Accountability
in AI Systems
Trust and Accountability
in AI Systems
Advancing verifiable and transparent AI with federated training, secure aggregation, confidential computing, and differential privacy for accountable and compliant finance.
Advancing verifiable and transparent AI with federated training, secure aggregation, confidential computing, and differential privacy for accountable and compliant finance.
Methodologies for Processing Unstructured Data
Techniques to construct structured data from unstructured financial data using LLMs and NLP.
Technical Challenges of LLMs and NLP
Addressing issues such as hallucination, interpretability, and the integration of NLP methods in financial applications.
Prompt Engineering
Optimizing LLM prompts to extract relevant financial insights.
Technical Challenges of LLMs and NLP
Addressing issues such as hallucination, interpretability, and the integration of NLP methods in financial applications.
Prompt Engineering
Optimizing LLM prompts to extract relevant financial insights.
Retrieval-Augmented Generation (RAG)
Integrating external data sources with LLMs and NLP techniques for enhanced information retrieval.
Fine-Tuning LLMs
Retrieval-Augmented Generation (RAG)
Customizing LLMs for specific financial applications to improve accuracy and relevance.
Integrating external data sources with LLMs and NLP techniques for enhanced information retrieval.
Text-to-SQL Applications
Text-to-SQL Applications
Translating natural language queries into SQL for querying structured databases using LLMs and NLP.
Translating natural language queries into SQL for querying structured databases using LLMs and NLP.
Methodologies for Processing Unstructured Data
Techniques to construct structured data from unstructured financial data using LLMs and NLP.
Evaluation Techniques
Evaluation Techniques
Enhancing the verifiability of generated text by LLMs and NLP methods to streamline manual verification processes.
Enhancing the verifiability of generated text by LLMs and NLP methods to streamline manual verification processes.
Applications in Analyzing Financial Reports and Alternative Data
Applications in Analyzing Financial Reports and Alternative Data
Utilizing LLMs and NLP for analyzing financial documents, news, SEC filings, and alternative data such as social media and video content.
Utilizing LLMs and NLP for analyzing financial documents, news, SEC filings, and alternative data such as social media and video content.
Financial Modeling
Financial Modeling
Applying LLMs and AI for credit assessment, price forecasting, bankruptcy detection, and other financial modeling scenarios.
Applying LLMs and AI for credit assessment, price forecasting, bankruptcy detection, and other financial modeling scenarios.
Multi-Lingual ESG Identification and Assessment
Multi-Lingual ESG Identification and Assessment
everaging LLMs and NLP for automated ESG scoring and identifying ESG issues across languages.
everaging LLMs and NLP for automated ESG scoring and identifying ESG issues across languages.
Financial Fraud Detection
Financial Fraud Detection
Using NLP approaches, including those involving LLMs, for detecting fraud in financial transactions.
Using NLP approaches, including those involving LLMs, for detecting fraud in financial transactions.
Enhancing Investor Communication
Enhancing Investor Communication
Utilizing LLMs and NLP to improve investor relations and communication.
Utilizing LLMs and NLP to improve investor relations and communication.
Submission
Guidelines
There is no length limit or strict format requirement. While formatting according to ACM guidelines is recommended (see below), we encourage submissions in an open format to allow the authors to focus on content with formatting restrictions. Submissions should be in PDF format and can be directly submitted as they are.
The workshop is non-archival and will not have official proceedings.
Submission Guidelines
There is no length limit or strict format requirement. While formatting according to ACM
guidelines is recommended (see below), we encourage submissions in an open format to
allow the authors to focus on content with formatting restrictions. Submissions should be in
PDF format and can be directly submitted as they are.
The workshop is non-archival and will not have official proceedings.
Recommended Formatting:
Additional Requirements
For each submission, at least one author must agree to serve as a reviewer and deliver their reviews on time.
The author list cannot be modified after the initial submission, except before printing.
Any changes after acceptance will be communicated directly with the authors.
For each submission, at least one author must agree to serve as
a reviewer and deliver their reviews on time.
The author list cannot be modified after the initial submission, except
before printing.Any changes after acceptance will be communicated directly with
the authors.
Key Dates
Submission Deadline
October 1st, 2025 (Anywhere on Earth)
Author Notification
October 15th, 2025 (Anywhere on Earth)
Workshop
November 15th, 2025 (Full Day Session)
Call for Papers
Submit
Submission Process
Submissions are to be made through the OpenReview platform. Authors must submit the paper content (PDF document), title, author names, contact details, and a brief abstract electronically through the workshop’s submission site. At least one author of each accepted paper is required to attend the conference to present their work.
Submission Process
Submissions are to be made through the OpenReview platform. Authors must submit the paper content (PDF document), title, author names, contact details, and a brief abstract electronically through the workshop’s submission site. At least one author of each accepted paper is required to attend the conference to present their work.
Review Process
This workshop will follow a single-blind review process, where authors’ identities are known to the reviewers, but reviewers’ identities are not disclosed to the authors. There will be no rebuttal period, and all submissions will be treated with strict confidentiality.
Additional Requirements
For each submission, at least one author must agree to serve as a reviewer and deliver their reviews on time.
The author list cannot be modified after the initial submission, except
before printing.Any changes after acceptance will be communicated directly with
the authors.
Accepted Papers
All accepted papers will be invited to participate in the poster session. Participants are required to print and bring their own posters to the event.
Detailed specifications on poster format and requirements will be provided by the organizers after acceptance.
Selected papers may also be given the opportunity for an oral presentation, subject to scheduling constraints.
Please note that the workshop is non-archival and will not have official proceedings. Only the author names and titles of accepted papers will appear on the website, with no disclosure of any content.
The workshop is non-archival and will not have official proceedings. Only the author names and titles of accepted papers will appear on the website, with no disclosure of any content.
Key Dates
Submission Deadline
October 1st, 2025 (Anywhere on Earth)
Author Notification
October 15th, 2025 (Anywhere on Earth)
Workshop
November 15th, 2025 (Full Day Session)
Call for Papers
Submit
Topics:
Include but not limited to:
•Exploring how multi-agent systems can collaborate to streamline decision-making, analysis, and research across diverse financial tasks.
•Retrieval-augmented generation grounded in Financial Corpora
•Evaluation under high-stakes and regulatory constraints
•Composable system design for explainable reasoning
•Customizing LLMs for specific financial applications to improve accuracy and relevance.
•Utilizing LLMs and NLP to improve investor relations and communication.
•Leveraging LLMs and NLP for automated ESG scoring and identifying ESG issues across languages.
•Translating natural language queries into SQL for querying structured databases using LLMs and NLP.
Submission Process
Submissions are to be made through the OpenReview platform. Authors must submit the paper content (PDF document), title, author names, contact details, and a brief abstract electronically through the workshop’s submission site. At least one author of each accepted paper is required to attend the conference to present their work.
Review Process
This workshop will follow a single-blind review process, where authors’ identities are known to the reviewers, but reviewers’ identities are not disclosed to the authors. There will be no rebuttal period, and all submissions will be treated with strict confidentiality.
Additional Requirements
For each submission, at least one author must agree to serve as a reviewer and deliver their reviews on time.
The author list cannot be modified after the initial submission, except
before printing.Any changes after acceptance will be communicated directly with
the authors.
Accepted Papers
All accepted papers will be invited to participate in the poster session. Participants are required to print and bring their own posters to the event.
Detailed specifications on poster format and requirements will be provided by the organizers after acceptance.
Selected papers may also be given the opportunity for an oral presentation, subject to scheduling constraints.
Please note that the workshop is non-archival and will not have official proceedings. Only the author names and titles of accepted papers will appear on the website, with no disclosure of any content.
The workshop is non-archival and will not have official proceedings. Only the author names and titles of accepted papers will appear on the website, with no disclosure of any content.
Recommended Formatting:
Organized by Members of the Following Institutions
Contact
For further details and submission guidelines, please contact
contact@ai4f.org