arXiv cs.CL· Xuan Zhao, Haonan He, Qingyu Yang, Minglei Li, Zelin Tan, Bo Wan, Jingqi Ye, Chen Zhang, Shuyue Hu, Peng Ye·· 12 小时前AI 评分40
ParametricSkills：将文本技能参数化为 LoRA 适配器
Parametric Skills
AI 导读
研究者提出 ParametricSkills 框架，在测试时将文本技能转换为 LoRA 适配器，实现无需占用上下文窗口的技能全生命周期管理。该框架先构建大规模高质量技能集合，合成单轮与多轮技能使用轨迹训练超网络，再用 OpenCode 采集的复杂长程轨迹做对齐。评估显示其显著提升任务完成度，并让技能以适配器形式累积，为测试时持续学习提供初步路径。
正文 · 原文
Abstract:For LLMs, skills are textual recipes encoding mature problem-solving experience. Despite widespread deployment, their utility is limited by the model's ability to comprehend and follow skill instructions. To address this limitation, we propose ParametricSkills, a framework that parameterizes textual skills at test time, enabling context-free management of the full skill lifecycle. Specifically, we first construct a large-scale and high-quality skill collection, and then synthesize single-turn and multi-turn skill-exploitation trajectories. Using this data, we then train a hypernetwork that parameterizes both the content and the exploitation methodologies of skills by converting textual skills into LoRA adapters. To facilitate the real-world application of our framework, we further align the hypernetwork on complex, long-horizon trajectories collected with OpenCode, enabling the models to operate reliably in realistic interactive workflows with ParametricSkills. Across evaluations, ParametricSkills effectively parameterizes skills, yielding substantial gains in task completion. More notably, as parametric skills accumulate as adapters rather than crowding the context, they open a preliminary but promising path toward continual learning at test time.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2606.30015 [cs.CL] |
| (or arXiv:2606.30015v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2606.30015 arXiv-issued DOI via DataCite |
Submission history
From: Haonan He [view email]
[v1]
Mon, 29 Jun 2026 09:19:32 UTC (1,382 KB)
[v2]
Thu, 5 Nov 2026 18:18:18 UTC (3,221 KB)
来源：arXiv cs.CL · arxiv.org