NeLLCom-Lex: A Neural-agent Framework to Study the Interplay between Lexical Systems and Language Use

2025-09-26

AI Summary

Existing observational methods cannot establish causal mechanisms underlying lexical semantic change, while human experiments are ill-suited for studying long-term diachronic evolution. Method: We propose a language-game simulation framework grounded in neural agents: starting from real-world lexical systems, we iteratively evolve multi-generational agent populations under controlled communicative pressures (e.g., discrimination demands in color-naming tasks), integrating supervised and reinforcement learning. Contribution/Results: This is the first framework to enable *causally interpretable* semantic evolution *from empirically grounded lexical initial states*. Experiments successfully replicate hallmark properties of human color terminology—including basic color categories—and generate human-like lexical structure within a single generation. Crucially, the system dynamically adapts its lexicon in response to manipulated communicative demands, providing an interventionist, reproducible, and mechanistically transparent computational validation pathway for semantic change.

Technology Category

• Natural Language Processing: Lexical Semantics and Morphology
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• 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
• Web Mining and Content Analysis: Models for Web evolution
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Lexical semantic change has primarily been investigated with observational and experimental methods; however, observational methods (corpus analysis, distributional semantic modeling) cannot get at causal mechanisms, and experimental paradigms with humans are hard to apply to semantic change due to the extended diachronic processes involved. This work introduces NeLLCom-Lex, a neural-agent framework designed to simulate semantic change by first grounding agents in a real lexical system (e.g. English) and then systematically manipulating their communicative needs. Using a well-established color naming task, we simulate the evolution of a lexical system within a single generation, and study which factors lead agents to: (i) develop human-like naming behavior and lexicons, and (ii) change their behavior and lexicons according to their communicative needs. Our experiments with different supervised and reinforcement learning pipelines show that neural agents trained to 'speak' an existing language can reproduce human-like patterns in color naming to a remarkable extent, supporting the further use of NeLLCom-Lex to elucidate the mechanisms of semantic change.

Problem

• Simulating semantic change using neural agents
• Studying causal mechanisms behind lexical evolution
• Investigating factors influencing human-like naming behavior

Innovation

• Simulates semantic change using neural-agent framework
• Grounds agents in real lexical systems for communication
• Trains agents with supervised and reinforcement learning