Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn how to transform an open-source model into a high-performing finance-specific large language model through this 16-minute conference talk from the AI Engineer World's Fair. Discover the comprehensive approach used to develop Albatross, a specialized LLM that achieves top leaderboard performance on both general chat and domain-specific financial tasks. Explore the intensive data pipeline and training regimen that combines continual pre-training, fine-tuning, and preference optimization to customize models for complex financial workflows including tabular reasoning and sentiment analysis. Understand the critical challenge of establishing reliable p99 performance levels necessary for financial agents to successfully execute chained tasks, and gain insights into overcoming execution hurdles that often cause AI projects to fail in specialized domains. Examine why most existing finance domain-specific LLMs rely solely on pre-training or supervised fine-tuning approaches, and learn how a multi-faceted training strategy can achieve superior results for financial applications.
