SkillGen: Learning Domain Skills for In-Context Sequential Decision Making

• URL: http://arxiv.org/abs/2511.14670v1
• Date: Tue, 18 Nov 2025 17:09:21 GMT
• Title: SkillGen: Learning Domain Skills for In-Context Sequential Decision Making
• Authors: Ruomeng Ding, Wei Cheng, Minglai Shao, Chen Zhao,
• Abstract summary: We introduce SkillGen, a skill-based ICL framework for structured sequential reasoning.<n>We show that SkillGen achieves consistent gains, improving progress rate by 5.9%-16.5% on average across models.
• Score: 24.41349550520032
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Beyond Description: Cognitively Benchmarking Fine-Grained Action for Embodied Agents [52.14392337070763]
  We introduce CFG-Bench, a new benchmark designed to systematically evaluate fine-grained action intelligence.<n>CFG-Bench consists of 1,368 curated videos paired with 19,562 three-modalities question-answer pairs targeting four cognitive abilities.<n>Our comprehensive evaluation on CFG-Bench reveals that leading MLLMs struggle to produce detailed instructions for physical interactions.
  arXiv  Detail & Related papers  (2025-11-24T02:02:29Z)
• Search-R3: Unifying Reasoning and Embedding Generation in Large Language Models [11.39711340224126]
  Search-R3 is a novel framework that adapts Large Language Models to generate search embeddings as a direct output of their reasoning process.<n>Our approach exploits LLMs' chain-of-thought capabilities, allowing them to produce more effective embeddings by reasoning step-by-step through complex semantic analyses.
  arXiv  Detail & Related papers  (2025-10-08T14:16:20Z)
• Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective [65.12150411762273]
  We show that pruning random demonstrations into seemingly incoherent "gibberish" can remarkably improve performance across diverse tasks.<n>We propose a self-discover prompt optimization framework, PromptQuine, that automatically searches for the pruning strategy by itself using only low-data regimes.
  arXiv  Detail & Related papers  (2025-06-22T07:53:07Z)
• Truly Assessing Fluid Intelligence of Large Language Models through Dynamic Reasoning Evaluation [106.17986469245302]
  Large language models (LLMs) have demonstrated impressive reasoning capacities that mirror human-like thinking.<n>Existing reasoning benchmarks either focus on domain-specific knowledge (crystallized intelligence) or lack interpretability.<n>We propose DRE-Bench, a dynamic reasoning evaluation benchmark grounded in a hierarchical cognitive framework.
  arXiv  Detail & Related papers  (2025-06-03T09:01:08Z)
• Real-Time Verification of Embodied Reasoning for Generative Skill Acquisition [47.068088124436535]
  Generative skill acquisition enables embodied agents to actively learn a scalable and evolving repertoire of control skills.<n>We propose VERGSA, a framework that systematically integrates real-time verification principles into embodied skill learning.<n>To the best of our knowledge, this approach constitutes the first comprehensive training dataset for verification-driven generative skill acquisition.
  arXiv  Detail & Related papers  (2025-05-16T12:19:13Z)
• CoT-RAG: Integrating Chain of Thought and Retrieval-Augmented Generation to Enhance Reasoning in Large Language Models [15.560280546809457]
  Chain-of-thought (CoT) reasoning boosts large language models' (LLMs) performance on complex tasks.<n>We propose CoT-RAG, a novel reasoning framework with three key designs.<n>We show significant accuracy gains-ranging from 4.0% to 44.3%-over state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-04-18T07:55:09Z)
• Dancing with Critiques: Enhancing LLM Reasoning with Stepwise Natural Language Self-Critique [66.94905631175209]
  We propose a novel inference-time scaling approach -- stepwise natural language self-critique (PANEL)<n>It employs self-generated natural language critiques as feedback to guide the step-level search process.<n>This approach bypasses the need for task-specific verifiers and the associated training overhead.
  arXiv  Detail & Related papers  (2025-03-21T17:59:55Z)
• GIVE: Structured Reasoning of Large Language Models with Knowledge Graph Inspired Veracity Extrapolation [108.2008975785364]
  Graph Inspired Veracity Extrapolation (GIVE) is a novel reasoning method that merges parametric and non-parametric memories to improve accurate reasoning with minimal external input.<n>GIVE guides the LLM agent to select the most pertinent expert data (observe), engage in query-specific divergent thinking (reflect), and then synthesize this information to produce the final output (speak)
  arXiv  Detail & Related papers  (2024-10-11T03:05:06Z)
• Benchmarking General-Purpose In-Context Learning [19.40952728849431]
  In-context learning (ICL) empowers generative models to address new tasks effectively and efficiently on the fly. In this paper, we study extending ICL to address a broader range of tasks with an extended learning horizon and higher improvement potential. We introduce two benchmarks specifically crafted to train and evaluate GPICL functionalities.
  arXiv  Detail & Related papers  (2024-05-27T14:50:42Z)
• Knowledgeable In-Context Tuning: Exploring and Exploiting Factual Knowledge for In-Context Learning [37.22349652230841]
  Large language models (LLMs) enable in-context learning (ICL) by conditioning on a few labeled training examples as a text-based prompt. In this paper, we demonstrate that factual knowledge is imperative for the performance of ICL in three core facets. We introduce a novel Knowledgeable In-Context Tuning (KICT) framework to further improve the performance of ICL.
  arXiv  Detail & Related papers  (2023-09-26T09:06:39Z)
• Improving Open Information Extraction with Large Language Models: A Study on Demonstration Uncertainty [52.72790059506241]
  Open Information Extraction (OIE) task aims at extracting structured facts from unstructured text. Despite the potential of large language models (LLMs) like ChatGPT as a general task solver, they lag behind state-of-the-art (supervised) methods in OIE tasks.
  arXiv  Detail & Related papers  (2023-09-07T01:35:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.