update.

rl-introduction.bbl

-\begin{thebibliography}{143}
+\begin{thebibliography}{144}
 \providecommand{\natexlab}[1]{#1}
 \providecommand{\url}[1]{\texttt{#1}}
 \expandafter\ifx\csname urlstyle\endcsname\relax
@@ -631,6 +631,11 @@ Mengzhou Xia, Sadhika Malladi, Suchin Gururangan, Sanjeev Arora, and Danqi Chen.
 \newblock Less: Selecting influential data for targeted instruction tuning.
 \newblock \emph{arXiv preprint arXiv:2402.04333}, 2024.
 
+\bibitem[Xia et~al.(2026)Xia, Chen, Wang, Liu, Zeng, Wang, Han, Zhou, Zhao, Chen, et~al.]{xia-etal:skillrl}
+Peng Xia, Jianwen Chen, Hanyang Wang, Jiaqi Liu, Kaide Zeng, Yu~Wang, Siwei Han, Yiyang Zhou, Xujiang Zhao, Haifeng Chen, et~al.
+\newblock Skillrl: Evolving agents via recursive skill-augmented reinforcement learning.
+\newblock \emph{arXiv preprint arXiv:2602.08234}, 2026.
+
 \bibitem[Xiao et~al.(2024)Xiao, Yuan, Zhu, Li, and Honavar]{xiao:2024cal}
 Teng Xiao, Yige Yuan, Huaisheng Zhu, Mingxiao Li, and Vasant~G. Honavar.
 \newblock Cal-dpo: Calibrated direct preference optimization for language model alignment.

rl-introduction.bib

@@ -4,7 +4,12 @@
 
 
 
-
+@article{xia-etal:skillrl,
+  title={Skillrl: Evolving agents via recursive skill-augmented reinforcement learning},
+  author={Xia, Peng and Chen, Jianwen and Wang, Hanyang and Liu, Jiaqi and Zeng, Kaide and Wang, Yu and Han, Siwei and Zhou, Yiyang and Zhao, Xujiang and Chen, Haifeng and others},
+  journal={arXiv preprint arXiv:2602.08234},
+  year={2026}
+}
 
 @article{yu-etal:skill,
   title={Skill is Not One-Size-Fits-All: Model-Aware Skill Alignment for LLM Agents},

section6/section6.tex

@@ -436,7 +436,7 @@ One limitation of manual skill construction is that the quality and diversity la
 
 The above way of generating skills often suffer from quality issues. First, a general LLM does not inherently know how to design effective skills without additional optimization. Second, the skill generator may optimize for the wrong objective because skill generation and skill utilization involve different contexts. Specifically, an LLM may consider a generated skill effective because it satisfies the instructions provided in the prompt, while another agent may fail to use this skill effectively in real-world tasks. As a result, skill generation should not only focus on producing well-structured skills, but also consider whether these skills can actually improve agent performance during task execution. Recent work resorts to RL to achieve this goal, where the skill generation is optimized based on the actual performance of generated skills.
 
-
+Here we consider \textbf{SkillRL} as an example to illustrate how to optimize the skill generation though RL \citep{xia-etal:skillrl}.