Schedule

Invite Talk

Title: From Text to Alpha: Multi-Agent and LLM Behaviors in Financial Reasoning

Jacob Chanyeol Choi LinqAlpha

Alejandro Lopez-Lira University of Florida

Abstract: Large language models (LLMs) are increasingly deployed as both analytical and trading agents in financial markets. This talk explores how these models read, reason, and act on financial information such as corporate disclosures and news articles. We present complementary perspectives: empirical evidence on the predictive power and limitations of LLM-based signals and market reactions, and experimental analysis of model-specific biases and reasoning behaviors that shape investment outcomes. Compared with traditional machine learning and textual models, these systems demonstrate substantially greater capacity to capture context, adapt to evolving signals, and surface latent structures in financial narratives. This session will illustrate how LLMs transition from text understanding to market participation, revealing both hidden biases and hidden alpha in financial reasoning.

Invite Talk

Title: LLMs Are Not Accountants: Data Without Disclosure via Rule‑Grounded World Models for Financial Ledgers

Furong Huang University of Maryland

Abstract: Financial AI is starved of real transaction data, and unconstrained LLMs hallucinate money long before they can replace it. PersonaLedger offers a way out: a rule-regulated world model where LLMs generate rich, persona-driven spending behavior while a programmatic engine enforces accounting reality. The result is a 30M-event synthetic ledger and a benchmark suite that finally lets us evaluate GenAI models for credit risk and fraud, without touching private data.

Invite Talk

Title: New Frontiers of Generative AI in Finance: Structured Agentic Workflows and foundational Time-Series Models

Hao Ni University College London

Abstract: Building accurate and trustworthy models for financial time-series data remains a core challenge in quantitative finance. This talk presents two complementary research directions advancing the frontiers of generative AI in finance: structured agentic workflows and domain-adapted foundation models.

In the first part, I will introduce TS-Agent, a modular framework that integrates the systematic optimization of AutoML with the reasoning and adaptability of agentic AI. TS-Agent automates time-series modelling through iterative stages of model selection, code refinement, and fine-tuning, guided by curated knowledge banks. Combining the strengths of both paradigms, it enables adaptive, transparent, and auditable model development, consistently outperforming strong AutoML and agentic baselines across diverse forecasting and generative tasks.

In the second part, I will present our empirical study on time-series foundation models (TSFMs) on financial data. Using a comprehensive dataset of daily excess returns across global markets, we systematically evaluate zero-shot inference, fine-tuning, and pre-training from scratch. Pre-training on financial data and synthetic data augmentation yields substantial gains in predictive accuracy and portfolio performance, underscoring the promise of TSFM in finance.

Invite Talk

Title: Continuous Simulation Loops - The Future of Agentic Quant Research

Ioana Boier Nvidia

Invite Talk

Title: Building Trustworthy AI for investment process

Stefano  Pasquali Domyn