需求驅動智能體訓練任務合成框架問世

Computer Science > Software Engineering arXiv:2607. 14186 (cs) [Submitted on 15 Jul 2026 (v1), last revised 22 Jul 2026 (this version, v4)] Title:NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs Authors:Jiarong Zhao, Zhikai Lei, Zhiheng Xi, Rui Zheng, Hang Yan, Jie Zhou, Qin Chen, Liang He View a PDF of the paper titled NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs, by Jiarong Zhao and 7 other authors View PDF Abstract:Scaling executable agent training data for LLM post-training is bottlenecked by substrate-bound methods that tie task generation to predefined tools, repositories, or skill graphs: expanding coverage requires manual substrate engineering, each new domain demands a bespoke pipeline, and the resulting task distributions often reflect substrate biases rather than real-world demand. We introduce NexForge, a requirement-driven framework that takes high-level capability requirements as input and synthesizes diverse, executable agent tasks and expert trajectories for SFT. NexForge first investigates real-world demand to construct representative scenarios and task profiles, then performs distribution-aware compilation to generate task directives. For each directive, NexForge automatically retrieves or constructs the required files, dependencies, and runtime configurations, and finally synthesizes expert rollouts and produces training trajectories. Without domain-specific infrastructure, NexForge produces 3. 6K terminal and 2K office tasks, improving Qwen3. 5-35B-A3B Base from 22. 0\% on Terminal-Bench 2.