模糊、符号与语境：通过认知支架增强大语言模型指令遵循能力
Fuzzy, Symbolic, and Contextual: Enhancing LLM Instruction via Cognitive Scaffolding
AI总结：
该研究探究提示层归纳偏置对大语言模型教学对话认知行为的影响，提出搭配短期记忆模式的符号支架方法，经消融实验验证完整系统优于基线，证明认知支架可有效塑造LLM的教学策略。
摘要：
我们研究提示层面的归纳偏置如何影响大语言模型（LLM）在教学对话中的认知行为。我们提出了一种符号支架方法，搭配短期记忆模式，旨在促进苏格拉底式辅导中的自适应结构化推理。我们通过在五个系统变体上进行受控消融实验，采用专家设计的评分标准评估模型输出，评分维度涵盖支架搭建、响应性、符号推理和对话记忆。我们展示了使用基于LLM的评估框架得到的初步结果，该框架与认知导向的评分标准对齐，可在早期实验阶段对不同架构变体进行可扩展的系统性比较。初步结果显示，我们的完整系统始终优于基线变体。分析表明，移除记忆或符号结构会削弱关键认知行为，包括抽象能力、自适应探询和概念连续性。这些发现支持了一种加工层面的解释：提示层面的认知支架能够可靠地塑造LLM中涌现的教学策略。
英文摘要：
We study how prompt-level inductive biases influence the cognitive behavior of large language models (LLMs) in instructional dialogue. We introduce a symbolic scaffolding method paired with a short-term memory schema designed to promote adaptive, structured reasoning in Socratic tutoring. Using controlled ablation across five system variants, we evaluate model outputs via expert-designed rubrics covering scaffolding, responsiveness, symbolic reasoning, and conversational memory. We present preliminary results using an LLM-based evaluation framework aligned to a cognitively grounded rubric. This enables scalable, systematic comparisons across architectural variants in early-stage experimentation. The preliminary results show that our full system consistently outperforms baseline variants. Analysis reveals that removing memory or symbolic structure degrades key cognitive behaviors, including abstraction, adaptive probing, and conceptual continuity. These findings support a processing-level account in which prompt-level cognitive scaffolds can reliably shape emergent instructional strategies in LLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
