小模型通过元学习掌握逻辑推理，低数据下表现超越大模型。
Teaching Small Language Models to Learn Logic through Meta-Learning
- 用元学习让小模型从少量样本中提炼通用逻辑规则
- 1.5B-7B模型在少样本下推理准确率显著提升
- 适合想用小模型实现强逻辑能力的研究者
大型语言模型（LLMs）在推理任务上的表现日益受到关注，但其逻辑能力仍存争议。为此，我们聚焦于形式逻辑中的一个明确片段——三段论推理，将问题建模为前提选择，并构建受控数据集以隔离逻辑能力。除了评估，一个开放挑战是使模型能够习得可泛化的抽象推理模式。本文提出采用少样本元学习方法，促使模型在任务间提取规则而非在任务内记忆模式。尽管元学习在逻辑可学习性领域研究较少，实验表明其效果显著：经过元学习微调的小模型（1.5B-7B）在泛化能力上表现优异，尤其在低数据场景下优势明显。这些元学习模型在我们的三段论推理任务上超越了GPT-4o和o3-mini。
原文摘要 · Abstract (English)
Large language models (LLMs) are increasingly evaluated on reasoning tasks, yet their logical abilities remain contested. To address this, we study LLMs' reasoning in a well-defined fragment of logic: syllogistic reasoning. We cast the problem as premise selection and construct controlled datasets to isolate logical competence. Beyond evaluation, an open challenge is enabling LLMs to acquire abstract inference patterns that generalize to novel structures. We propose to apply few-shot meta-learning to this domain, thereby encouraging models to extract rules across tasks rather than memorize patterns within tasks. Although meta-learning has been little explored in the context of logic learnability, our experiments show that it is effective: small models (1.5B-7B) fine-tuned with meta-learning demonstrate strong gains in generalization, with especially pronounced benefits in low-data regimes. These meta-learned models outperform GPT-4o and o3-mini on our syllogistic reasoning task.
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。