{"ID":2840146,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.14670","arxiv_id":"2511.14670","title":"SkillGen: Learning Domain Skills for In-Context Sequential Decision Making","abstract":"Large language models (LLMs) are increasingly applied to sequential decision-making through in-context learning (ICL), yet their effectiveness is highly sensitive to prompt quality. Effective prompts should meet three principles: focus on decision-critical information, provide step-level granularity, and minimize reliance on expert annotations through label efficiency. However, existing ICL methods often fail to satisfy all three criteria simultaneously. Motivated by these challenges, we introduce SkillGen, a skill-based ICL framework for structured sequential reasoning. It constructs an action-centric, domain-level graph from sampled trajectories, identifies high-utility actions via temporal-difference credit assignment, and retrieves step-wise skills to generate fine-grained, context-aware prompts. We further present a theoretical analysis showing that focusing on high-utility segments supports task identifiability and informs more effective ICL prompt design. Experiments on ALFWorld, BabyAI, and ScienceWorld, using both open-source and proprietary LLMs, show that SkillGen achieves consistent gains, improving progress rate by 5.9%-16.5% on average across models.","short_abstract":"Large language models (LLMs) are increasingly applied to sequential decision-making through in-context learning (ICL), yet their effectiveness is highly sensitive to prompt quality. Effective prompts should meet three principles: focus on decision-critical information, provide step-level granularity, and minimize relia...","url_abs":"https://arxiv.org/abs/2511.14670","url_pdf":"https://arxiv.org/pdf/2511.14670v1","authors":"[\"Ruomeng Ding\",\"Wei Cheng\",\"Minglai Shao\",\"Chen Zhao\"]","published":"2025-11-18T17:09:21Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}