Analogical Structure, Minimal Contextual Cues and Contrastive Distractors: Input Design for Sample-Efficient Linguistic Rule Induction

2025-11-13

AI Summary

This study addresses the challenge of few-shot linguistic rule induction. We propose a cognitively inspired analogical paradigm for input organization, grounded in three principles: analogical structural modeling, minimal contextual cue extraction, and contrastive distractor design—implemented as a structured cloze task. Using a lightweight BERT+CNN model (0.5M parameters), our approach achieves F1 = 0.95 with only ~100 training examples, substantially outperforming zero-shot GPT-4o (F1 = 0.87). Ablation studies confirm the efficacy of each design component, and cross-linguistic evaluations across diverse phenomena—including tense, number agreement, and word order—demonstrate strong robustness. Our results show that analogy-driven input organization significantly enhances the rule induction capability of lightweight models under extreme data scarcity, offering a new pathway toward efficient, interpretable language learning paradigms.

Technology Category

• Cognitive Modeling & Cognitive Systems: Analogy
• Natural Language Processing: Lexical Semantics and Morphology
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Large language models achieve strong performance through training on vast datasets. Can analogical paradigm organization enable lightweight models to match this performance with minimal data? We develop a computational approach implementing three cognitive-inspired principles: analogical structure, contrastive learning, and minimal contextual cues. We test this approach with structured completion tasks where models identify correct sentence completions from analogical patterns with contrastive alternatives. Training lightweight models (BERT+CNN, $0.5M$ parameters) on only one hundred structured examples of English causative/inchoative alternations achieves $F1=0.95$, outperforming zero-shot exttt{GPT-o3} ($F1=0.87$). Ablation studies confirm that analogical organization and contrastive structure improve performance, consistently surpassing randomly shuffled baselines across architectures. Cross-phenomenon validation using unspecified object alternations replicates these efficiency gains, confirming approach robustness. Our results show that analogical paradigm organization enables competitive linguistic rule learning with orders of magnitude less data than conventional approaches require.

Problem

• Enabling lightweight models to learn linguistic rules with minimal training data
• Testing cognitive-inspired principles for efficient analogical pattern recognition
• Achieving high performance with small models using structured analogical paradigms

Innovation

• Using analogical structure for efficient learning
• Implementing contrastive learning with minimal cues
• Training lightweight models with small datasets