Uncertainty in LLM Explanations - Understanding Uncertainty from Reasoning Topology

Explore uncertainty in large language model explanations through reasoning topology analysis in this 17-minute conference talk from METACOG-25. Learn about Longchao Da's research from Arizona State University examining how uncertainty manifests in LLM-generated explanations and the implications for understanding model reasoning processes. Discover methodologies for analyzing explanation reliability and gain insights into the intersection of symbolic reasoning and deep learning approaches to artificial intelligence.