一作ACL 2026Findings· Findings of ACL
F use S earch: Learning Adaptive Parallel Execution for Efficient Code Localization
Ke Xu · Siyang Xiao · Ming Liang · Yichen Yu · Zhixiang Wang · Jingxuan Xu · Dajun Chen · Wei Jiang · Yong Li
摘要
Code localization is a primary bottleneck in automated software development. While parallel tool execution can accelerate discovery, existing agents suffer from a 34.9% redundant tool invocation rate, negating the benefits of parallelism. We introduce FuseSearch, which reframes parallel code loca...