SWE-MILE: Asynchronous Potential-Induced Milestone Credit Assignment for Long-Horizon Software Engineering Agents
Long-horizon software engineering (SWE) agents trained with reinforcement learning with verifiable rewards (RLVR) typically receive only terminal outcome supervision, making it difficult to distinguish productive actions from redundant exploration or functional regressions. We propose SWE-MILE, an asynchronous potentia...
Chao-Qun Cui, Hao Zhou, Mei-Qi Chen et al.
· 0 citations