Schemex: Interactive Structural Abstraction from Examples with Contrastive Refinement

2025-04-16

AI Summary

Inducing implicit structural schemas from diverse, heterogeneous instances remains a challenging cognitive task for users. Method: We propose Schemex, the first interactive schema abstraction framework, which operationalizes schema induction—grounded in cognitive science—as an interpretable, human-AI collaborative modeling process. It integrates hierarchical clustering, visual abstraction mapping, contrastive refinement, and an AI-assisted reasoning interface to support users in identifying deep structural commonalities beneath surface-level variations. Contribution/Results: Unlike end-to-end AI approaches, Schemex enables iterative, user-driven clustering, abstraction, and feedback-driven refinement. A user study demonstrates that Schemex significantly improves structural insight and modeling confidence over AI-only baselines (p < 0.01), empirically validating the efficacy of adaptive, explainable human-AI co-abstraction.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Each type of creative or communicative work is underpinned by an implicit structure. People learn these structures from examples - a process known in cognitive science as schema induction. However, inducing schemas is challenging, as structural patterns are often obscured by surface-level variation. We present Schemex, an interactive visual workflow that scaffolds schema induction through clustering, abstraction, and contrastive refinement. Schemex supports users through visual representations and interactive exploration that connect abstract structures to concrete examples, promoting transparency, adaptability, and effective human-AI collaboration. In our user study, participants reported significantly greater insight and confidence in the schemas developed with Schemex compared to those created using a baseline of an AI reasoning model. We conclude by discussing the broader implications of structural abstraction and contrastive refinement across domains.

Problem

• Extracting implicit structures from varied examples
• Supporting schema induction via interactive visualization
• Enhancing human-AI collaboration in structural abstraction

Innovation

• Interactive visual workflow for schema induction
• Clustering and abstraction to reveal structures
• Contrastive refinement for human-AI collaboration