Composing or Not Composing? Towards Distributional Construction Grammars

2024-12-10

arXiv.org

AI Summary

Language understanding faces the fundamental challenge of reconciling compositional and non-compositional mechanisms. This paper proposes Distributional Construction Grammar (DCG), the first formal framework unifying distributional semantics with construction grammar. DCG employs extended feature structures to uniformly represent constructions, semantic frames, and events, and models meaning composition as an activation process governed by semantic similarity and unification constraints. The framework simultaneously supports compositional derivation and the interpretation of non-compositional phenomena—including idioms, metaphors, and conventionalized expressions—while ensuring cognitive plausibility and data-driven learnability. Empirical evaluation demonstrates DCG’s effectiveness in modeling diverse non-compositional expressions across multiple linguistic domains. By integrating symbolic structure with distributional representations, DCG establishes a novel, computationally tractable paradigm for language understanding that is both formally rigorous and cognitively interpretable.

Technology Category

• Natural Language Processing: Lexical Semantics and Morphology
• Knowledge Representation and Reasoning: Description Logics
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

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

Abstract

The mechanisms of comprehension during language processing remains an open question. Classically, building the meaning of a linguistic utterance is said to be incremental, step-by-step, based on a compositional process. However, many different works have shown for a long time that non-compositional phenomena are also at work. It is therefore necessary to propose a framework bringing together both approaches. We present in this paper an approach based on Construction Grammars and completing this framework in order to account for these different mechanisms. We propose first a formal definition of this framework by completing the feature structure representation proposed in Sign-Based Construction Grammars. In a second step, we present a general representation of the meaning based on the interaction of constructions, frames and events. This framework opens the door to a processing mechanism for building the meaning based on the notion of activation evaluated in terms of similarity and unification. This new approach integrates features from distributional semantics into the constructionist framework, leading to what we call Distributional Construction Grammars.

Problem

• Modeling compositional and non-compositional language comprehension mechanisms
• Integrating construction grammar with distributional semantics approaches
• Developing unified framework for meaning representation through construction interactions

Innovation

• Integrating distributional semantics into construction grammar framework
• Formalizing feature structure representation in Sign-Based Construction Grammars
• Using activation-based processing with similarity and unification