用符号模型约束大模型，让步骤解释更讲逻辑。
Improving Procedural Skill Explanations via Constrained Generation: A Symbolic-LLM Hybrid Architecture
- 符号结构约束大模型生成过程
- 在'如何'和'为何'问题上提升解释质量
- 适合教育类智能辅导系统使用
在程序性技能学习中，教学解释不仅需说明步骤，还需传达其背后的因果关系、目标导向和组合逻辑。大语言模型（LLM）常生成流畅但浅层的回复，缺乏这种结构。我们提出Ivy——一种人工智能教练系统，通过结合符号化的任务-方法-知识（TMK）模型与生成式解释层（即受约束的LLM），实现结构化、多步骤的解释。TMK编码因果转换、目标层级与问题分解，并为LLM生成提供明确的结构边界。我们在三个推断维度上，通过专家与独立标注评估Ivy与GPT及检索增强GPT基线的表现。结果表明，符号约束能持续提升对'如何'和'为何'问题的解释结构质量。本研究展示了一种可扩展的教育类AI方法，增强了智能辅导系统中AI生成解释的教学价值。
原文摘要 · Abstract (English)
In procedural skill learning, instructional explanations must convey not just steps, but the causal, goal-directed, and compositional logic behind them. Large language models (LLMs) often produce fluent yet shallow responses that miss this structure. We present Ivy, an AI coaching system that delivers structured, multi-step explanations by combining symbolic Task-Method-Knowledge (TMK) models with a generative interpretation layer-an LLM that constructs explanations while being constrained by TMK structure. TMK encodes causal transitions, goal hierarchies, and problem decompositions, and guides the LLM within explicit structural bounds. We evaluate Ivy against responses against GPT and retrieval-augmented GPT baselines using expert and independent annotations across three inferential dimensions. Results show that symbolic constraints consistently improve the structural quality of explanations for "how" and "why" questions. This study demonstrates a scalable AI for education approach that strengthens the pedagogical value of AI-generated explanations in intelligent coaching systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。