Reading Between the Lines: Modeling User Behavior and Costs in AI-Assisted Programming
Name
3613904.3641936.pdf
Size
982.38 KB
Format
Adobe PDF
Checksum (MD5)
2f6912341f5a6d1f9303bdefabf97566
Author(s)
Mozannar, Hussein
Bansal, Gagan
Fourney, Adam
Horvitz, Eric
Date Issued
May 11, 2024
Publisher
ACM
Citation
Mozannar, Hussein, Bansal, Gagan, Fourney, Adam and Horvitz, Eric. 2024. "Reading Between the Lines: Modeling User Behavior and Costs in AI-Assisted Programming."
Version
Final published version
Abstract
Code-recommendation systems, such as Copilot and CodeWhisperer, have the potential to improve programmer productivity by suggesting and auto-completing code. However, to fully realize their potential, we must understand how programmers interact with these systems and identify ways to improve that interaction. To seek insights about human-AI collaboration with code recommendations systems, we studied GitHub Copilot, a code-recommendation system used by millions of programmers daily. We developed CUPS, a taxonomy of common programmer activities when interacting with Copilot. Our study of 21 programmers, who completed coding tasks and retrospectively labeled their sessions with CUPS, showed that CUPS can help us understand how programmers interact with code-recommendation systems, revealing inefficiencies and time costs. Our insights reveal how programmers interact with Copilot and motivate new interface designs and metrics.
Description
CHI '24: Proceedings of the CHI Conference on Human Factors in Computing Systems May 11–16, 2024, Honolulu, HI, USA
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3613904.3641936
