通过原子技能扩大编码代理
Scaling Coding Agents via Atomic Skills
AI总结：
本文提出通过原子技能提升编码代理的新型方法，通过联合强化学习提升五个基础技能，从而提高复杂软件任务的泛化能力。
摘要：
当前LLM编码代理主要在复合基准上训练，导致任务特定过拟合和泛化能力有限。为此，我们提出一种新的扩展范式，将重点从任务级优化转向原子技能掌握。我们首先正式化五个基本原子技能，即代码定位、代码编辑、单元测试生成、问题重现和代码审查，这些技能作为复杂软件工程任务的基础向量。与复合编码任务相比，这些原子技能更具通用性和可组合性。然后，我们通过联合强化学习扩展编码代理，使原子技能一致提升，而不会产生负面影响或权衡。值得注意的是，这些原子技能的改进在其他未见的复合编码任务中表现良好，如bug修复、代码重构、机器学习工程和代码安全。观察到这一现象，促使我们通过训练原子技能提出新的编码代理扩展范式。广泛实验验证了所提范式的有效性。值得注意的是，我们的联合强化学习在5个原子技能和5个复合任务上平均性能提高了18.7%。
英文摘要：
Current LLM coding agents are predominantly trained on composite benchmarks (e.g., bug fixing), which often leads to task-specific overfitting and limited generalization. To address this, we propose a novel scaling paradigm that shifts the focus from task-level optimization to atomic skill mastery. We first formalize five fundamental atomic skills, code localization, code editing, unit-test generation, issue reproduction, and code review, that serve as the basis vectors for complex software engineering tasks. Compared with composite coding tasks, these atomic skills are more generalizable and composable. Then, we scale coding agents by performing joint RL over atomic skills. In this manner, atomic skills are consistently improved without negative interference or trade-offs between them. Notably, we observe that improvements in these atomic skills generalize well to other unseen composite coding tasks, such as bug-fixing, code refactoring, machine learning engineering, and code security. The observation motivates a new scaling paradigm for coding agents by training with atomic skills. Extensive experiments demonstrate the effectiveness of our proposed paradigm. Notably, our joint RL improves average performance by 18.7% on 5 atomic skills and 5 composite tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
