The Dual Nature of Consistency in Foundation Models - Challenges and Opportunities
Toronto Machine Learning Series (TMLS) via YouTube
Overview
Watch a 36-minute conference talk from the Toronto Machine Learning Series exploring the complex relationship between consistency and foundation models, particularly Large Language Models (LLMs). Learn from AI Risk and Vulnerability Alliance Science Lead Jekaterina Novikova as she delves into the measurement methods for model consistency, discusses strategies to mitigate negative impacts, and reveals opportunities to leverage consistency advantageously. Gain valuable insights into the critical role consistency plays in developing trustworthy AI models and understand both its challenges and potential benefits in the evolving landscape of foundation models.
Syllabus
The Dual Nature of Consistency in Foundation Models: Challenges and Opportunities
Taught by
Toronto Machine Learning Series (TMLS)