Agent-based models for the evolution of morphological alternation patterns
Summary
This paper presents multi-agent simulations of the emergence of morphological alternation patterns (like 'go/went') in language, using an AI Historical Linguist (LLM-driven) to evaluate plausibility of evolved morphologies against real languages.
View Cached Full Text
Cached at: 06/12/26, 08:50 AM
# Agent-based models for the evolution of morphological alternation patterns Source: [https://arxiv.org/abs/2606.12748](https://arxiv.org/abs/2606.12748) [View PDF](https://arxiv.org/pdf/2606.12748) > Abstract:Why is the past of English "go" the apparently unrelated "went"? Such alternations are frequent in languages\. They neither aid communication nor learnability, yet they can be persistent, surviving over centuries or millennia\. We present a multi\-agent simulation of the emergence of morphological stem and inflection alternations\. Alternate forms arise by phonological changes or, as with "go/went", from lexical alternatives associated with a subset of the population\. When an agent 'hears' another agent use a novel form for a slot in the paradigm of a word \(say, the past tense of go\), they will with some probability adopt that form, possibly spreading its use to other slots in the paradigm that shared the same original form\. Thus alternative forms can spread through the population and become entrenched as stem or inflectional marker alternants\. Unlike many previous computational studies, our system allows for naturalistic lexical forms, realistic phonological rules, lexicons with hundreds or thousands of entries, and agent populations in the tens or hundreds\. It supports several network topologies, diffusion patterns and agent adoption policies\. One issue with such simulations is evaluation: how realistic is the resulting morphology compared to those of real languages? We introduce the AI Historical Linguist, a novel Large Language Model\-driven system that models a debate between two historical linguists\. We use this to compare a set of real language morphologies, disguised morphologies, and experimentally evolved morphologies\. The results suggest that among the factors that favor more plausible morphologies are scale\-free social networks and random Bernoulli adoption of forms\. We also present three case studies modeling attested historical changes, allowing us to test what might have happened if history had been different\. All code and data are released\. ## Submission history From: Richard Sproat \[[view email](https://arxiv.org/show-email/3a297006/2606.12748)\] **\[v1\]**Wed, 10 Jun 2026 23:26:44 UTC \(5,283 KB\)
Similar Articles
Emergent Languages in Populations of Language Model Agents: From Token Efficiency to Oversight Evasion
This paper studies emergent languages that autonomous LLM agents propose to one another on the Moltbook platform, finding that some languages are specifically designed to evade human oversight and can be learned in-context from short descriptions. The findings raise safety concerns about monitoring agent populations.
HawkesLLM: Semantic Uncertainty Propagation in Agentic Text Simulation
This paper introduces HawkesLLM, a framework that models semantic uncertainty propagation in multi-step agentic text simulations by combining a multivariate Hawkes process for temporal influence and memory selection with a language model for text generation. Evaluation on a GDELT news-cascade case study shows improved late-stage semantic alignment under compact prompt-memory constraints.
@Skoorbkaz: BREAKING REPORT: New research involving @AnthropicAI researcher Jack Lindsey and collaborators has demonstrated somethi…
New research demonstrates that natural language 'mind viruses' can evolve and spread between AI agents through persistent memory, altering behavior and posing a real but limited risk in multi-agent LLM systems, as detailed in a paper published on arXiv.
Exploring Collaboration between a language and a non-language agent
This paper introduces LLAMIA-Bench and a method called latent state internalization to enhance collaboration between language models and non-language agents, showing that internalizing continuous representations outperforms verbalization and matches frontier models like GPT-5.1.
From BERT to Frontier Agents: Eight Years of Language-Model Progress, the Collapse of the Capability-Cost Curve, and the Rise of Task-Targeted Models
This paper reviews eight years of language model evolution from BERT to advanced agentic systems, documenting the collapse of the capability-cost curve and the emergence of task-targeted models through empirical benchmarks and reproducible experiments.