Workshop on AI in Finance

Welcome to the 2nd edition of the Workshop in AI and Finance at ECAI-2025.

The importance of Artificial Intelligence has grown a lot in the last decade with companies of different sizes starting their own AI research labs. Recently, this phenomenon started to take place also in the financial service industry with large financial corporations equipping themselves with research departments to appropriately leverage AI in the context of their business. Therefore, this workshop aims to bring together researchers and practitioners both from industry and academia in order to share and discuss recent advancements in the context of AI applications for financial services. The intended target audience is represented by all AI researchers actively investigating ways to leverage AI in finance, researchers and practitioners that would like to explore potential applications of their work in this domain, and professionals looking for a deeper understanding of the potential application AI has in the context of their business.

This workshop is related to a series of events at the intersection of AI in Finance: ICLR Workshop on Advances in Financial AI, AI for Social Impact, KDD Finance Day.

Previous year's workshop can be found here: WAIFin2024

Important Dates

• Submissions due:  May 23 2025 23:59 AoE

• Notification to authors: July 11 2025 23:59 AoE

• Workshop date: October 26 2025

Registration

To participate in the workshop, please register via the ECAI website. Please note that selecting the weekend option is sufficient for attenting solely in the workshop.

Invited Talk

Talk: Reinforcement Learning in Algorithmic Trading (slides)

Abstract: 

Reinforcement Learning (RL) provides a general framework for sequential decision making based on interaction and feedback. Although RL has demonstrated remarkable success in areas such as games and natural language processing, its application to financial markets introduces distinctive challenges, including nonstationarity, risk aversion, delayed feedback, and control frequency. This talk presents the formulation of trading as a reinforcement learning problem and outlines recent methodological advances developed at the RL³ (Real Life Reinforcement Learning) Lab of Politecnico di Milano. The proposed approaches address market nonstationarity through adaptive transfer techniques, incorporate risk-aware optimization using coherent risk measures, and enable effective learning under temporal delays and continuous-time dynamics. By connecting theoretical developments with practical trading scenarios, the presentation highlights how reinforcement learning can contribute to more robust, adaptive, and interpretable algorithmic trading strategies.

Accepted Papers

Accepted papers will be presented either as oral presentations or as posters. The maximum poster size allowed is A0.

• A ML dataflow to monitor ESG-related news articles and detect corporate greenwashing. Luca Montalto-Giampaoli, Ilaria Peri and Alessandro Provetti. (ORAL PRESENTATION)

• Agentic NLP for Thematic Analysis of Fed Speeches: Yield Impacts under Jerome Powell (2018–2025). Akshar Prabhu Desai. (POSTER)
  
  

• CAESar: Conditional Autoregressive Expected Shortfall. Federico Gatta, Fabrizio Lillo and Piero Mazzarisi. (ORAL PRESENTATION)
  
  

• Can News Predict the Direction of Oil Price Volatility? A Language Model Approach with SHAP Explanations. Felipe Maldonado and Romina Hashami (ORAL PRESENTATION)
  
  

• Causal Discovery for Stochastic Power Law Time Series. Matteo Tusoni, Giuseppe Masi, Andrea Coletta, Aldo Glielmo, Viviana Arrigoni and Novella Bartolini. (POSTER)
  
  

• Cryptocurrencies in the Balance Sheet: Insights from (Micro)Strategy - Bitcoin Interactions. Sabrina Aufiero, Antonio Briola, Tesfaye Salarin, Silvia Bartolucci, Fabio Caccioli and Tomaso Aste. (ORAL PRESENTATION)
  
  

• Data-Driven Strategy for Merchant Incentive Optimization in Digital Payment Ecosystems. Hira Ajmal. (POSTER)
  
  

• Datachat: a multi-agent assistant for domain specific tasks. Silvia Pagliarini, Lorenzo Bonetti, Marco Colognesi and Fabio Tesser. (POSTER)
  
  

• Graph Neural Networks for Financial Time Series Forecasting: A Multi-Modal Edge Construction Approach. Jiuyu Zhang and Silvia Bartolucci. (POSTER)
  
  

• Hierarchical Topic Modeling and New Intent Discovery with LLMs.  Aaron Rodrigues, Mahmood Hegazy and Azzam Naeem. (ORAL PRESENTATION)
  
  

• How does AI understand Central Bank communication? Chiara Lecchini, Oliver Giudice and Andrea Coletta. (POSTER)
  
  

• Human-machine integration in stock market prediction. Alberto Matuozzo, Paul Yoo and Alessandro Provetti. (POSTER)
  
  

• MAFA: A Multi-Agent Framework for Annotation. Mahmood Hegazy, Aaron Rodrigues and Azzam Naeem. (ORAL PRESENTATION)
  
  

• Methodological Insights into Structural Causal Modelling and Uncertainty-Aware Forecasting for Economic Indicators. Federico Cerutti. (POSTER)
  
  

• Monetary Policy Forecasting from Central Bank Communications: An Embedding-Based Approach. Yucheng Lu and Man Chon Iao. (POSTER)
  
  

• Prompting for Policy: Interpreting Monetary Policy Signals with Synthetic LLM Personas. Giulia Iadisernia and Carolina Camassa. (ORAL PRESENTATION)
  
  

• Reinforcement Learning and Agent-Based Models for Execution Algorithm Optimisation. Ollie Olby, Rory Baggott and Namid Stillman. (ORAL PRESENTATION)
  
  

• Reinforcement Learning for Optimal Execution when Liquidity is Time-Varying. Andrea Macri and Fabrizio Lillo. (ORAL PRESENTATION)
  
  

• Reinforcement Learning for Optimal Execution with the Queue Reactive Model. Tomas Espana, Yadh Hasi, Edoardo Vittori and Fabrizio Lillo. (ORAL PRESENTATION)
  
  

• Sentiment Analysis vs. techical indicators in the forecasting of niche commodity markets returns. Gabriella Anita Herke, Alberto Matuozzo, Alessandro Provetti and Paul Yoo. (POSTER)
  
  

• Synthetic Query Generation for Expanding Intent Recognition. Aaron Rodrigues, Mahmood Hegazy and Azzam Naeem. (POSTER)
  
  

• TABL-ABM: A Hybrid Framework for Synthetic LOB Generation. Ollie Olby, Rory Baggott and Namid Stillman. (ORAL PRESENTATION)
  
  

• The Liquidity Game: Strategic Coordination and Market Design in Bilateral Trading. Alicia Vidler and Toby Walsh. (ORAL PRESENTATION)
  
  

• Topic Labeling Using Large Language Models. Olga Bogachek. (POSTER)
  
  

• Unlocking NACE Classification Embeddings with OpenAI for Enhanced Analysis and Processing. Andrea Vidali, Nicola Jean and Giacomo Le Pera. (POSTER)

Industry Panel

• Daniele Regoli - Intesa Sanpaolo

• Andrea Coletta - Banca D'Italia

• Namid Stillman - Simudyne

Venue

The workshop will be held at the Engineering School of the University of Bologna: Viale del Risorgimento 2, 40136, Bologna, Italy.

Organizers

Program Commitee

• Alessio Brini 

• Andrea Coletta

• Anne Rolim

• Anton Esmail-Yakas

• Antonio Briola

• Brian Huge

• Federico Cacciamani

• Fernando Acero

• Giovanni Dispoto

• Giulia Preti

• Junhyeong Lee

• Keane Wei Yang Ong

• Lorenzo Lucchese

• Maria Katsidoniotaki

• Meghana Holla

• Michele Trapletti

• Nazanin Mehrasa

• Rasmus Jensen

• Saeid Amiri

• Santiago Marro

• Tomas España

• Uljad Berdica

• Yadh Hafsi

• Zheng Wang

Sponsors