Preprint
Jul 2026
SWE-NFI: Studying and Benchmarking Coding Agents for Non-Functional Improvements
This paper operationalizes developer-oriented NFIs into 92 executable rules and develops a comprehensive evaluation suite that combines functional correctness testing with rule-based NFI evaluation, and evaluates state-of-the-art commercial and open-source coding agents.
Pengyu Xue, Hengchen Yuan, Xin Wang et al.
· 0 citations