1 paper · 1 filter
Bowen Ye, Rang Li, Qibin Yang +10
Large language models are increasingly deployed as autonomous agents for multi-step workflows in real-world software environments. However, existing agent benchmarks are limited by…