Training Versatile Coding Agents in Synthetic Environments

• URL: http://arxiv.org/abs/2512.12216v1
• Date: Sat, 13 Dec 2025 07:02:28 GMT
• Title: Training Versatile Coding Agents in Synthetic Environments
• Authors: Yiqi Zhu, Apurva Gandhi, Graham Neubig,
• Abstract summary: We introduce SWE-Playground, a novel pipeline for generating environments and trajectories.<n>SWE-Playground synthetically generates projects and tasks from scratch with strong language models and agents.<n>This allows us to tackle a much wider variety of coding tasks, such as reproducing issues by generating unit tests and implementing libraries from scratch.
• Score: 44.5849223659282
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Prior works on training software engineering agents have explored utilizing existing resources such as issues on GitHub repositories to construct software engineering tasks and corresponding test suites. These approaches face two key limitations: (1) their reliance on pre-existing GitHub repositories offers limited flexibility, and (2) their primary focus on issue resolution tasks restricts their applicability to the much wider variety of tasks a software engineer must handle. To overcome these challenges, we introduce SWE-Playground, a novel pipeline for generating environments and trajectories which supports the training of versatile coding agents. Unlike prior efforts, SWE-Playground synthetically generates projects and tasks from scratch with strong language models and agents, eliminating reliance on external data sources. This allows us to tackle a much wider variety of coding tasks, such as reproducing issues by generating unit tests and implementing libraries from scratch. We demonstrate the effectiveness of this approach on three distinct benchmarks, and results indicate that SWE-Playground produces trajectories with dense training signal, enabling agents to reach comparable performance with significantly fewer trajectories than previous works.

Related papers

• BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing? [61.247730037229815]
  We introduce BeyondSWE, a comprehensive benchmark that broadens existing evaluations along two axes - resolution scope and knowledge scope.<n>To investigate the role of external knowledge, we develop SearchSWE, a framework that integrates deep search with coding abilities.<n>This work offers both a realistic, challenging evaluation benchmark and a flexible framework to advance research toward more capable code agents.
  arXiv  Detail & Related papers  (2026-03-03T17:52:01Z)
• Hybrid-Gym: Training Coding Agents to Generalize Across Tasks [59.95803522351185]
  In this paper, we describe some transferable skills that are shared across diverse tasks.<n>We propose a training environment, Hybrid-Gym, consisting of a set of scalable synthetic tasks.<n>Experiments show that agents trained on our synthetic tasks effectively generalize diverse real-world tasks.
  arXiv  Detail & Related papers  (2026-02-18T19:30:55Z)
• ASTRA: Automated Synthesis of agentic Trajectories and Reinforcement Arenas [13.919124676472022]
  ASTRA is an end-to-end framework for training tool-augmented language model agents.<n>ASTRA integrates scalable data synthesis and verifiable reinforcement learning.<n> Experiments on multiple agentic tool-use benchmarks demonstrate that ASTRA-trained models achieve state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-01-29T11:22:23Z)
• Software Engineering Agents for Embodied Controller Generation : A Study in Minigrid Environments [3.415592919976024]
  Software Engineering Agents (SWE-Agents) have proven effective for traditional software engineering tasks with accessibles.<n>We present the first extended evaluation of SWE-Agents on controller generation for embodied tasks.
  arXiv  Detail & Related papers  (2025-10-24T16:04:11Z)
• Reinforcement Learning for Machine Learning Engineering Agents [52.03168614623642]
  We show that agents backed by weaker models that improve via reinforcement learning can outperform agents backed by much larger, but static models.<n>We propose duration- aware gradient updates in a distributed asynchronous RL framework to amplify high-cost but high-reward actions.<n>We also propose environment instrumentation to offer partial credit, distinguishing almost-correct programs from those that fail early.
  arXiv  Detail & Related papers  (2025-09-01T18:04:10Z)
• InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling [71.37579508777843]
  Large language models (LLMs) have revolutionized artificial intelligence by enabling complex reasoning capabilities.<n>To address this gap, we present InternBootcamp, an open-source framework comprising 1000+ domain-diverse task environments.
  arXiv  Detail & Related papers  (2025-08-12T05:00:00Z)
• A Survey on Code Generation with LLM-based Agents [61.474191493322415]
  Code generation agents powered by large language models (LLMs) are revolutionizing the software development paradigm.<n>LLMs are characterized by three core features.<n>This paper presents a systematic survey of the field of LLM-based code generation agents.
  arXiv  Detail & Related papers  (2025-07-31T18:17:36Z)
• SWE-rebench: An Automated Pipeline for Task Collection and Decontaminated Evaluation of Software Engineering Agents [31.921127664873882]
  LLM-based agents have shown promising capabilities in a growing range of software engineering (SWE) tasks.<n>High-quality training data is scarce, especially data that reflects real-world SWE scenarios.<n>Existing datasets are either limited to one-shot code generation or comprise small, manually curated collections of interactive tasks.
  arXiv  Detail & Related papers  (2025-05-26T18:01:00Z)
• Spatial Reasoning and Planning for Deep Embodied Agents [2.7195102129095003]
  This thesis explores the development of data-driven techniques for spatial reasoning and planning tasks. It focuses on enhancing learning efficiency, interpretability, and transferability across novel scenarios.
  arXiv  Detail & Related papers  (2024-09-28T23:05:56Z)
• Environment Generation for Zero-Shot Compositional Reinforcement Learning [105.35258025210862]
  Compositional Design of Environments (CoDE) trains a Generator agent to automatically build a series of compositional tasks tailored to the agent's current skill level. We learn to generate environments composed of multiple pages or rooms, and train RL agents capable of completing wide-range of complex tasks in those environments. CoDE yields 4x higher success rate than the strongest baseline, and demonstrates strong performance of real websites learned on 3500 primitive tasks.
  arXiv  Detail & Related papers  (2022-01-21T21:35:01Z)
• Multi-Task Learning with Sequence-Conditioned Transporter Networks [67.57293592529517]
  We aim to solve multi-task learning through the lens of sequence-conditioning and weighted sampling. We propose a new suite of benchmark aimed at compositional tasks, MultiRavens, which allows defining custom task combinations. Second, we propose a vision-based end-to-end system architecture, Sequence-Conditioned Transporter Networks, which augments Goal-Conditioned Transporter Networks with sequence-conditioning and weighted sampling.
  arXiv  Detail & Related papers  (2021-09-15T21:19:11Z)

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.