Traditional interventions for academic procrastination often fail due to their inability to address individual-specific factors. Large language models (LLMs) offer potential for personalized interventions, but user expectations and LLM limitations are underexplored. This study used interviews and focus groups with university students and experts to evaluate a technology probe generating personalized procrastination advice. Results highlight the need for structured, deadline-oriented steps, enhanced user support, and adaptive questioning based on factors like busyness. Findings offer design implications for LLM-based procrastination management tools while cautioning against using LLMs for therapeutic guidance.

Ananya Bhattacharjee, Yuchen Zeng, Sarah Yi Xu, Dana Kulzhabayeva, Minyi Ma, Rachel Kornfield, Syed Ishtiaque Ahmed, Alex Mariakakis, Mary P Czerwinski, Anastasia Kuzminykh, Michael Liut, Joseph Jay Williams