R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE Agents

• URL: http://arxiv.org/abs/2504.07164v1
• Date: Wed, 09 Apr 2025 17:55:19 GMT
• Title: R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE Agents
• Authors: Naman Jain, Jaskirat Singh, Manish Shetty, Liang Zheng, Koushik Sen, Ion Stoica,
• Abstract summary: AgentGym is the largest procedurally-curated executable gym environment for training real-world SWE-agents.<n>It is powered by two main contributions: SYNGEN, a synthetic data curation recipe, and Hybrid Test-time Scaling.<n>Overall, our approach achieves 51% on the SWE-Bench Verified benchmark, reflecting a new state-of-the-art for open-weight SWE-agents.
• Score: 32.06393076572057
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Improving open-source models on real-world SWE tasks (solving GITHUB issues) faces two key challenges: 1) scalable curation of execution environments to train these models, and, 2) optimal scaling of test-time compute. We introduce AgentGym, the largest procedurally-curated executable gym environment for training real-world SWE-agents, consisting of more than 8.7K tasks. AgentGym is powered by two main contributions: 1) SYNGEN: a synthetic data curation recipe that enables scalable curation of executable environments using test-generation and back-translation directly from commits, thereby reducing reliance on human-written issues or unit tests. We show that this enables more scalable training leading to pass@1 performance of 34.4% on SWE-Bench Verified benchmark with our 32B model. 2) Hybrid Test-time Scaling: we provide an in-depth analysis of two test-time scaling axes; execution-based and execution-free verifiers, demonstrating that they exhibit complementary strengths and limitations. Test-based verifiers suffer from low distinguishability, while execution-free verifiers are biased and often rely on stylistic features. Surprisingly, we find that while each approach individually saturates around 42-43%, significantly higher gains can be obtained by leveraging their complementary strengths. Overall, our approach achieves 51% on the SWE-Bench Verified benchmark, reflecting a new state-of-the-art for open-weight SWE-agents and for the first time showing competitive performance with proprietary models such as o1, o1-preview and sonnet-3.5-v2 (with tools). We will open-source our environments, models, and agent trajectories.

Related papers

• SWE-Synth: Synthesizing Verifiable Bug-Fix Data to Enable Large Language Models in Resolving Real-World Bugs [10.70881967278009]
  We present SWE- Synth, a framework for synthesizing realistic verifiable, and process-aware bug-fix datasets at the repository level. Compared to manually curated datasets, our method scales with minimal human effort while preserving contextual richness and correctness. Our results highlight the potential of synthetic, agent-generated data to advance the state of the art in APR and software engineering automation.
  arXiv  Detail & Related papers  (2025-04-20T22:37:43Z)
• Review, Refine, Repeat: Understanding Iterative Decoding of AI Agents with Dynamic Evaluation and Selection [71.92083784393418]
  Inference-time methods such as Best-of-N (BON) sampling offer a simple yet effective alternative to improve performance. We propose Iterative Agent Decoding (IAD) which combines iterative refinement with dynamic candidate evaluation and selection guided by a verifier.
  arXiv  Detail & Related papers  (2025-04-02T17:40:47Z)
• Thinking Longer, Not Larger: Enhancing Software Engineering Agents via Scaling Test-Time Compute [61.00662702026523]
  We propose a unified Test-Time Compute scaling framework that leverages increased inference-time instead of larger models.<n>Our framework incorporates two complementary strategies: internal TTC and external TTC.<n>We demonstrate our textbf32B model achieves a 46% issue resolution rate, surpassing significantly larger models such as DeepSeek R1 671B and OpenAI o1.
  arXiv  Detail & Related papers  (2025-03-31T07:31:32Z)
• Boosting Virtual Agent Learning and Reasoning: A Step-wise, Multi-dimensional, and Generalist Reward Model with Benchmark [72.46357004059661]
  We propose Similar, a step-wise Multi-dimensional Generalist Reward Model. It offers fine-grained signals for agent training and can choose better action for inference-time scaling. We introduce the first benchmark in the virtual agent domain for step-wise, multi-dimensional reward model training and evaluation.
  arXiv  Detail & Related papers  (2025-03-24T13:30:47Z)
• START: Self-taught Reasoner with Tools [51.38785489790888]
  We introduce START (Self-Taught Reasoner with Tools), a tool-integrated long Chain-of-thought (CoT) reasoning LLM. START is capable of performing complex computations, self-checking, exploring diverse methods, and self-ging. It significantly outperforms the base QwQ-32B and achieves performance comparable to the state-of-the-art open-weight model R1-Distill-Qwen-32B.
  arXiv  Detail & Related papers  (2025-03-06T17:11:51Z)
• S*: Test Time Scaling for Code Generation [55.11863577956177]
  We propose S*, the first hybrid test-time scaling framework for code generation. S* substantially improves the coverage and selection accuracy of generated code.
  arXiv  Detail & Related papers  (2025-02-20T09:18:53Z)
• Training Software Engineering Agents and Verifiers with SWE-Gym [89.55822534364727]
  SWE-Gym is the first environment for training real-world software engineering (SWE) agents.<n>SWE-Gym contains 2,438 real-world Python task instances.
  arXiv  Detail & Related papers  (2024-12-30T18:15:39Z)
• The BrowserGym Ecosystem for Web Agent Research [151.90034093362343]
  BrowserGym ecosystem addresses the growing need for efficient evaluation and benchmarking of web agents.<n>We propose an extended BrowserGym-based ecosystem for web agent research, which unifies existing benchmarks from the literature.<n>We conduct the first large-scale, multi-benchmark web agent experiment and compare the performance of 6 state-of-the-art LLMs across 6 popular web agent benchmarks.
  arXiv  Detail & Related papers  (2024-12-06T23:43:59Z)
• AutoPenBench: Benchmarking Generative Agents for Penetration Testing [42.681170697805726]
  This paper introduces AutoPenBench, an open benchmark for evaluating generative agents in automated penetration testing. We present a comprehensive framework that includes 33 tasks, each representing a vulnerable system that the agent has to attack. We show the benefits of AutoPenBench by testing two agent architectures: a fully autonomous and a semi-autonomous supporting human interaction.
  arXiv  Detail & Related papers  (2024-10-04T08:24:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.