paper-with-me

홈 › Papers

Generative Skill Composition for LLM Agents

2026-06-30 · Xinyu Zhao, Zhen Tan, Vaishnav Tadiparthi, Nakul Agarwal, Kwonjoon Lee, Ehsan Moradi Pari, Hossein Nourkhiz Mahjoub, Tianlong Chen arxiv

Recent LLM agents benefit from skills for solving complex tasks. Skills encapsulate modular packages of procedural knowledge and instructions for performing specialized tasks, such as setting up a sandboxed environment, running a test suite, or refactoring a function across multiple files. As skill libraries grow and become reusable across tasks and domains, selecting an appropriate skill composition has emerged as a central bottleneck. Existing approaches fall into two categories. One exposes the agent's reasoning to the entire skill collection; the other performs skill retrieval via embeddings or LLM-based rerankers. Both provide useful insights; however, they miss the structural nature of skill composition, which is a joint decision over which skills, how many, and in what order -- three dimensions that cannot be decoupled. We formalize this as structured skill composition: given a task and a skill library, predict an executable skill plan that jointly specifies the activated subset, count, and execution order. We propose SkillComposer, which instantiates structured skill composition as task-conditioned skill sequence prediction. SkillComposer uses a constrained autoregressive decoder over skill identifiers, so subset, count, and order emerge jointly from a single decoding pass, and dependencies between successive skills are captured naturally. We build a training set of task-composition pairs from a real, human-curated skill library. We then evaluate SkillComposer along two axes: composition quality on a held-out test set, and downstream task success on SkillsBench across two production-grade coding agents. On GPT-5.2-Codex, Gemini-3-Pro-Preview, SkillComposer raises the pass rate by +23.1, +18.2pp over the no-skill baseline, surpassing top-3 retrieval and matching the gold-skill retrieval upper bound at lower prompt-token cost.

📄 PDF Abstract BibTeX arXiv:2606.32025

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?

2026-02-28 · Shiqi Chen, Jingze Gai, Ruochen Zhou, Jinghan Zhang 외 arxiv

Real-world tool-using agents operate over long-horizon workflows with recurring structure and diverse demands, where effective behavior requires not only invoking atomic tools but also abstracting, and reusing higher-lev…

Self-supervised Visual Reinforcement Learning with Object-centric Representations

2020-11-29 · ICLR 2021 1 · Andrii Zadaianchuk, Maximilian Seitzer, Georg Martius

Autonomous agents need large repertoires of skills to act reasonably on new tasks that they have not seen before. However, acquiring these skills using only a stream of high-dimensional, unstructured, and unlabeled obser…

Objectreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Compositional Skill Routing for LLM Agents: Decompose, Retrieve, and Compose

2026-06-16 · Xueping Gao arxiv

LLM agents increasingly rely on external skills -- reusable tool specifications -- but real-world tasks often require composing multiple skills, not just selecting one. We formalize this as the Compositional Skill Routin…

CompoSkill: Compositional Skill Chain Attacks from Individually Scanner-Passing LLM Agent Skills

2026-08-17 · Mingxiao Liu, Zhoumian Jiang, Jianan Ma, Jian Zhang 외 arxiv

Autonomous AI agents tackling Long Horizon Tasks depend on marketplace skills that are certified one at a time: a scanner returns a safety verdict for each skill and declares the ecosystem safe if every package passes. W…

SkillFuzz: Fuzzing Skill Composition for Implicit Intents Discovery in Open Skill Marketplaces

2026-07-02 · Jinwei Hu, Yi Dong, Youcheng Sun, Xiaowei Huang arxiv

Large Language Model (LLM)-based agents increasingly automate software engineering tasks through reusable skills, natural-language instruction documents that guide planning and execution. Open skill marketplaces enable u…

Intent Discovery