My paper has been accepted by International Journal of Human-Computer Studies

• Do you need help? Identifying and responding to pilots’ troubleshooting through eye-tracking and Large Language Model
  In this study, we introduce an innovative approach that tokenizes eye-tracking data into Visual Attention Matrices (VAMs) and integrates them with Large Language Models to identify and respond to pilots’ troubleshooting activities in real-time. This represents one of the early works combining eye-tracking data with LLMs for adaptive aviation support systems. The method addresses two key challenges: capturing the complex troubleshooting behaviors of actively engaged (In-the-Loop) pilots, and effectively processing non-semantic eye-tracking data using LLM technology.