FC-SWE: Failure-Conditioned RL for Long-Horizon Software Engineering Agents
Repository-level software engineering (SWE) is a challenging long-horizon setting: agents must reason over extended interactions, use tools, and adapt to stateful environments. Recent work trains SWE agents with reinforcement learning methods such as Group Relative Policy Optimization (GRPO), which independently sample...
Liu-Fu Jia, Bin Hu, Lin Jing et al.
· 0 citations