Toward Training Superintelligent Software Agents through Self-Play SWE-RL
While current software agents powered by large language models (LLMs) and agentic reinforcement learning (RL) can boost programmer productivity, their training data (e.g., GitHub issues and pull requests) and environments (e.g., pass-to-pass and fail-to-pass tests) heavily depend on human knowledge or curation, posing a fundamental barrier to superintelligence. In this paper, we present Self-play SWE-RL (SSR), a first step toward training paradigms for superintelligent software agents. Our approach takes minimal data assumptions, only requiring access to sandboxed repositories with source code and installed dependencies, with no need for human-labeled issues or tests. Grounded in these real-world codebases, a single LLM agent is trained via reinforcement learning in a self-play setting to iteratively inject and repair software bugs of increasing complexity, with each bug formally specified by a test patch rather than a natural language issue description. On the SWE-bench Verified and SWE-Bench Pro benchmarks, SSR achieves notable self-improvement (+10.4 and +7.8 points, respectively) and consistently outperforms the human-data baseline over the entire training trajectory, despite being evaluated on natural language issues absent from self-play. Our results, albeit early, suggest a path where agents autonomously gather extensive learning experiences from real-world software repositories, ultimately enabling superintelligent systems that exceed human capabilities in understanding how systems are constructed, solving novel challenges, and autonomously creating new software from scratch.
Code (0)
등록된 구현이 없습니다.
Tasks
Reinforcement LearningSimilar Papers 제목 키워드 기반
Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
Large language models (LLMs) are becoming the foundation for autonomous agents that can use tools to solve complex tasks. Reinforcement learning (RL) has emerged as a common approach for injecting such agentic capabiliti…
Reinforcement LearningAn Anthropic Argument against the Future Existence of Superintelligent Artificial Intelligence
This paper uses anthropic reasoning to argue for a reduced likelihood that superintelligent AI will come into existence in the future. To make this argument, a new principle is introduced: the Super-Strong Self-Sampling …
SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from Experience
Repurposing large vision-language models (LVLMs) as computer use agents (CUAs) has led to substantial breakthroughs, primarily driven by human-labeled data. However, these models often struggle with novel and specialized…
Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?
Large Language Models (LLMs) are reshaping almost all industries, including software engineering. In recent years, a number of LLM agents have been proposed to solve real-world software problems. Such software agents are…
Don't Fear the Reaper: Refuting Bostrom's Superintelligence Argument
In recent years prominent intellectuals have raised ethical concerns about the consequences of artificial intelligence. One concern is that an autonomous agent might modify itself to become "superintelligent" and, in sup…