Search papers, labs, and topics across Lattice.
This paper introduces Mastermind, a dual-loop framework designed to enhance repository-level vulnerability reproduction in software engineering by focusing on strategy learning rather than full action trajectories. By employing a trainable planner that learns reusable strategies through supervised fine-tuning and milestone-based gradient-reinforced policy optimization, Mastermind achieves an impressive 84.5% pass rate on CyberGym tasks, significantly outperforming existing methods. The approach allows for the independent training of the planner while maintaining the performance of multiple frozen executors, showcasing the effectiveness of high-level strategy learning in improving SE agents.
Learning high-level strategies can boost vulnerability reproduction success rates by over 20%, revolutionizing how we approach software security tasks.
Repository-level vulnerability reproduction is a demanding software engineering (SE) task: an agent must inspect a codebase, infer the input grammar that reaches a vulnerable path, construct a proof-of-conceptv(PoC), and verify that the crash disappears on the patched build. Recent LLM agents can often execute these steps when the approach is correct, yet they still fail by choosing the wrong strategy. This paper argues that strategy, rather than the full action trajectory, is the right learning unit for such SE agents: it is compact enough to optimize, concrete enough to guide execution, and stable enough to store and reuse across attempts. We present Mastermind, a dual-loop framework that separates transferable strategy learning from task-specific experience. A trainable planner learns reusable vulnerability-reproduction strategies through SFT and milestone-based GRPO, while an experience loop maintains task-local strategy records that guide subsequent attempts. The planner is trained independently of the executor, allowing strategy learning to improve multiple frozen executors without modifying their action-generation capability. We evaluate Mastermind on CyberGym using 260 training tasks and 200 held-out evaluation tasks. With GPT-5.5 as the frozen executor, Mastermind achieves an 84.5% pass rate, outperforming open-book PoC context (60.0%), Best-of-8 sampling (63.0%), and iterative improvement (77.0%). The same planner also improves GPT-5.4 mini and GLM~5.1 from 45.0% and 58.5% to 60.0% and 71.0%. These results demonstrate that learning high-level strategies is an effective and transferable mechanism for improving repository-scale SE agents.
