RepoMirage: Probing Repository Context Reasoning in Code Agents with Perturbations
Original article: https://arxiv.org/abs/2605.26177v1
Code agents are currently having skillful performance on repository-level software engineering benchmarks, but it remains unclear whether success on end-to-end tasks such as issue resolution truly reflects repository con...
This entry is part of the Top 50 AI Agent Articles curation.
