Compositionality and the lexicon in evolutionary semantics

2026-06-25 · Source: Computation and Language · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Computational Linguistics · Depth: Expert, quick

Summary

A new framework integrates formal semantics' insight into recursive lexical meaning composition with evolutionary modeling. This framework allows lexical meanings and a composition function to co-evolve under pressures for conceptual simplicity and communicative accuracy. Applied to quantificational meaning, the analysis reveals that conservativity, a well-known semantic universal, emerges as an efficient system-wide abstraction. The approach considers syntactic structure and helps resolve discrepancies between empirical evidence on quantifier learnability and existing evolutionary models. More broadly, this work demonstrates the productive combination of formal semantics and evolutionary modeling, providing a template for investigating universals related to global compression within grammatical categories, semantic specialization of syntactic arguments, and the co-evolution of lexical and compositional meaning.

Key takeaway

For research scientists exploring language evolution or designing compositional AI systems, this framework offers a novel approach to understanding how semantic universals emerge. You should consider applying this co-evolutionary model to investigate how conceptual simplicity and communicative accuracy shape meaning systems. This can inform the development of more robust and interpretable AI models that mimic natural language's compositional properties and efficient abstractions like conservativity.

Key insights

The framework integrates formal semantics with evolutionary modeling to explain semantic universals like conservativity.

Principles

• Lexical meanings and composition co-evolve.
• Conservativity emerges as an efficient abstraction.

Method

The framework allows lexical meanings and a composition function to co-evolve under pressures for conceptual simplicity and communicative accuracy, analyzing the Pareto frontier.

In practice

• Model co-evolution of lexical and compositional meaning.
• Analyze semantic specialization of arguments.

Topics

• Formal Semantics
• Evolutionary Modeling
• Compositionality
• Semantic Universals
• Quantificational Meaning
• Conservativity

Best for: AI Scientist, Research Scientist

Related on AIssential

• Agent-based models for the evolution of morphological alternation patterns — Agent-based simulations, evaluated by an LLM-driven "AI Historical Linguist," model the emergence and spread of morphological alternations in language.
• The Evolution of Meaning: NLP’s Long March from Syntax to Semantics — Meaning is a dynamic process requiring active computation and external anchoring, not an inherent property of a word.
• The anti-Bayesian is standing at the back window with a shotgun, scanning for priors coming over the hill, while a million assumptions just walk right into his house through the front door. (also, an interesting point by Yann LeCun in 2012 about human language) — Human language evolved for brain decodability, and vision is a more complex cognitive task than language.
• Floating or Suggesting Ideas? A Large-Scale Contrastive Analysis of Metaphorical and Literal Verb-Object Constructions — Metaphorical and literal language differences are largely construction-specific, not universally consistent.
• The Dynamics of Human and AI-Generated Language: How Semantics Fluctuates across Different Timescales — The ACW-0 semantic timescale analysis effectively distinguishes temporal organization of generic versus specific content in human and AI speech.
• RAG's Evolution: From Simple Retrieval to Agentic AI — Search evolution progresses from keyword matching to semantic understanding and autonomous, adaptive retrieval.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computation and Language.