Abstract:Understanding the computational operations involved in conceptual composition is fundamental for theories of language. However, the existing literature on this topic remains fragmented, comprising disconnected theories from various fields. For instance, while formal semantic theories in Linguistics rely on type-driven interpretation without explicitly representing the conceptual content of lexical items, neurolinguistic research suggests that the brain is sensitive to conceptual factors during word composition. What is the relationship between these two types of theories? Do they describe two distinct aspects of composition, operating independently, or do they connect in some way during interpretation by our brain? To probe this, we explored how the mathematical operations explaining the combination of two words into a phrase are affected by the semantic content of items and the formal linguistic relations between the combining items. For six phrase types that varied properties relevant to type-driven interpretation such as modification vs. argument-saturation and modifier context sensitivity, we collected human ratings of experiential semantic features both for the phrases and for all the individual words within the phrases. We then compared the ability of different computational combination rules to explain the phrase ratings based on the individual word ratings. Our results indicate that composition operations are not one-size-fits-all but rather depend on both feature type and linguistic relation. For example, in intersective Adjective-Noun phrases, addition is used to merge attention-related features, while color features are predominantly determined by the first word's ratings. In the case of social features, the verb chiefly guides interpretation in Verb-Noun phrases, whereas in Noun-Noun phrases, the model employs multiplication to combine the social features of the nouns.

What Meaning-Form Correlation Has to Compose With

A Complexity-Based Theory of Compositionality

Compositionality in Context

Measuring Meaning Composition in the Human Brain with Composition Scores from Large Language Models

What makes Models Compositional? A Theoretical View: With Supplement

A probabilistic framework for analysing the compositionality of conceptual combinations

Compositionality and Contextuality: The Symbolic and Statistical Theories of Meaning

Unlocking the complexity of phrasal composition: An interplay between semantic features and linguistic relations

Geometry of Compositionality

Geometric Signatures of Compositionality Across a Language Model's Lifetime

The paradox of the compositionality of natural language: a neural machine translation case study

Modelling Compositionality and Structure Dependence in Natural Language

From Words to Worlds: Compositionality for Cognitive Architectures

Major Constituents and Two Dependency Grammar Constraints on Sharing in Coordination

Compositionality in a Parallel Architecture for Language Processing

Compositionality as Lexical Symmetry

The Validity of Evaluation Results: Assessing Concurrence Across Compositionality Benchmarks

Montague semantics and modifier consistency measurement in neural language models

Experimental Support for a Categorical Compositional Distributional Model of Meaning

Partial correlations in compositional data analysis

A corpus-based analysis of adjective-noun collocations in the academic writing of native and non-native speakers of English