This empirical study analyzes over 29,000 GitHub pull requests to uncover how AI agents differ from humans in code modification granularity and semantic alig...
Level: advanced
By Daniel Ogenrwot, John Businge
Category: research