Reading Between the Lines: Modeling User Behavior and Costs in AI-Assisted Programming

arXiv:2210.14306v5 Announce Type: replace-cross
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 …

abstract arxiv auto behavior code codewhisperer copilot costs cs.hc cs.lg cs.se however identify modeling productivity programmer programmers programming reading recommendation recommendation systems seek systems type