Parametric Skills
2026-06-29 • Computation and Language
Computation and Language
AI summaryⓘ
The authors focus on improving how large language models (LLMs) use skills, which are like step-by-step instructions for solving problems. They find that current models struggle to follow these skill instructions well, especially when the instructions are long or complicated. To fix this, the authors create ParametricSkills, a method that turns written skills into special parameters the model can use easily without needing the full text each time. They test this approach on software engineering tasks and show it works better than just giving the instructions directly, and it also helps the model learn continuously over time.
Large Language ModelsSkillsIn-context LearningParametricSkillsLoRA AdaptersHypernetworkSkill LibrarySoftware Engineering TasksContinual LearningDeepSeek
Authors
Xuan Zhao, Haonan He, Qingyu Yang, Minglei Li, Jingqi Ye, Zelin Tan, Bo Wan, Peng Ye
Abstract
Since intelligence fundamentally relies on efficient skill acquisition (Chollet, 2019), the ability to leverage skills is critical. For LLMs, skills, manually authored or extracted from task trajectories, are textual recipes encoding mature problem-solving experience and are critical to agentic capabilities. Despite widespread deployment, their utility is limited by the model's ability to comprehend and follow skill instructions, especially under complex and long-context scenarios, where key instructions are difficult to locate and adhere to. To address this limitation, we propose ParametricSkills, a framework that can convert free-form textual skills into parameters at test time, enabling context-free skill exploitation. Specifically, we first construct a large-scale, high-quality skill library, and synthesize single-turn and multi-turn skill exploitation trajectories built around these skills with OpenCode. Using these data, we then train a hypernetwork that parameterizes both the skill content and the test-time exploitation methodology by receiving textual skills and converting them into LoRA adapters. Experimental results on six complex software engineering (SWE) subtasks demonstrate that, the proposed ParametricSkills averagely outperforms in-context learning by 6.44 points as judged by DeepSeek-V4-Flash, while also achieving significantly higher BERT Score and F1 score, confirming its effectiveness. Beyond performance, we further find that parametric skills, being inherently accumulative, offer a preliminary yet promising avenue toward test-time continual learning.