用程序化生成自动构建可扩展的工具交互环境，提升大模型实战能力。
EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis
- 通过主题挖掘与逻辑建模自动生成多样化环境骨架
- 合成191个环境、约7000个任务场景，支持多轮多工具交互训练
- 适合想提升大模型复杂任务执行能力的研究者和开发者
大语言模型（LLMs）需在真实世界环境中作为智能体行动，但高质量的工具交互沙箱难以获取。真实系统访问受限，模拟环境易产生幻觉，手动构建又难扩展。本文提出EnvScaler，一个基于程序化合成的可扩展工具交互环境框架。该框架包含两个组件：首先，SkelBuilder通过主题挖掘、逻辑建模与质量评估生成多样化的环境骨架；其次，ScenGenerator为每个环境生成多任务场景及规则驱动的轨迹验证函数。利用EnvScaler，我们合成191个环境与约7000个场景，并用于Qwen3系列模型的监督微调（SFT）与强化学习（RL）。三个基准测试结果表明，经训练的模型在涉及多轮、多工具交互的复杂环境中任务解决能力显著提升。代码与数据已开源至https://github.com/RUC-NLPIR/EnvScaler。
原文摘要 · Abstract (English)
Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but this process relies on rich and varied tool-interaction sandboxes. However, access to real systems is often restricted; LLM-simulated environments are prone to hallucinations and inconsistencies; and manually built sandboxes are hard to scale. In this paper, we propose EnvScaler, an automated framework for scalable tool-interaction environments via programmatic synthesis. EnvScaler comprises two components. First, SkelBuilder constructs diverse environment skeletons through topic mining, logic modeling, and quality evaluation. Then, ScenGenerator generates multiple task scenarios and rule-based trajectory validation functions for each environment. With EnvScaler, we synthesize 191 environments and about 7K scenarios, and apply them to Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) for Qwen3 series models. Results on three benchmarks show that EnvScaler significantly improves LLMs' ability to solve tasks in complex environments involving multi-turn, multi-tool interactions. We release our code and data at https://github.com/RUC-NLPIR/EnvScaler.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。