الملخص
Human-written agent skills encode rich workflows for real-world problem solving, but are typically used as external inference-time instructions rather than internalized as reusable model capabilities. We introduce \texttt{SkillGym}, a framework that transforms these skills into executable, verifiable training environments for large language model agents. Its skill-to-task pipeline instantiates concrete tasks, verifies outcomes with code-based checkers, and assesses empirical skill dependence through contrastive executions. We construct and release 2,756 environments across 12 categories and collect 8,364 successful trajectories from multiple models and harnesses, averaging 49 tool calls and over 60k logged text tokens. These resources support supervised fine-tuning on verified workflows and reinforcement learning with outcome-based rewards. Under Claude Code, supervised fine-tuning improves Qwen3.5-35B-A3B by 199 Elo on GDPval-AA v2, 19.10 percentage points on Terminal-Bench 2.1, and 28.13 and 12.38 points on SkillsBench v1.1 with and without skills, respectively. Our 35B \texttt{SkillGym-Agent} reaches 51.47\% on skill-assisted SkillsBench, exceeding reported scores for Claude Sonnet 4.6, GPT-5.4 Mini, and DeepSeek V4 Pro. Without skills, it also surpasses skill-assisted bases under Codex and Claude Code, suggesting reusable procedural competence.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Ge, Z., Shao, Y., Yang, Y., Cai, Y., Zhou, J., Chen, K., Zhang, B., Chen, Q., & He, L. (2026). SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving. https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving
MLA 9
Ge, Zhilong, et al. "SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving." https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving.
شيكاغو (المؤلف–التاريخ)
Ge, Zhilong, Yuting Shao, Yutao Yang, Yuxuan Cai, Jie Zhou, Kai Chen, Bo Zhang, Qin Chen, and Liang He. 2026. "SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving." https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving.
هارفارد
Ge, Z., Shao, Y., Yang, Y., Cai, Y., Zhou, J., Chen, K., Zhang, B., Chen, Q. and He, L. (2026) 'SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving', Available at: https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving.
فانكوفر
Ge Z, Shao Y, Yang Y, Cai Y, Zhou J, Chen K, et al. SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving. https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving
IEEE
Z. Ge, Y. Shao, Y. Yang, Y. Cai, J. Zhou, K. Chen, B. Zhang, Q. Chen, and L. He, "SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving," https://omanscience.com/ar/articles/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving.