Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Edward L., Kiasari, Mohammad Sharifi, Wang, Tianyu, Helm, Hayden, Athreya, Avanti, Priebe, Carey, Lyzinski, Vince
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2506.00077
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911565928726528
author Wang, Edward L.
Kiasari, Mohammad Sharifi
Wang, Tianyu
Helm, Hayden
Athreya, Avanti
Priebe, Carey
Lyzinski, Vince
author_facet Wang, Edward L.
Kiasari, Mohammad Sharifi
Wang, Tianyu
Helm, Hayden
Athreya, Avanti
Priebe, Carey
Lyzinski, Vince
contents Large language models (LLMs) are powerful tools that, in a number of settings, overlap with the results of human pattern recognition and reasoning. Retrieval-augmented generation (RAG) further allows LLMs to produce tailored output depending on the contents of their RAG databases. However, LLMs depend on complex, computationally expensive algorithms. In this paper, we introduce interacting Gaussian mixture models (GMMs) as a proxy for interacting LLMs. We construct a model of interacting GMMs, complete with an analogue to RAG updating, under which GMMs can generate, exchange, and update data and parameters. We show that this interacting system of Gaussian mixture models, which can be implemented at minimal computational cost, mimics certain aspects of experimental simulations of interacting LLMs whose iterative responses depend on feedback from other LLMs. We build a Markov chain from this system of interacting GMMs; formalize and interpret the notion of polarization for such a chain; and prove lower bounds on the probability of polarization. This provides theoretical insight into the use of interacting Gaussian mixture models as a computationally efficient proxy for interacting large language models.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00077
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Gaussian mixture models as a proxy for interacting language models
Wang, Edward L.
Kiasari, Mohammad Sharifi
Wang, Tianyu
Helm, Hayden
Athreya, Avanti
Priebe, Carey
Lyzinski, Vince
Computation and Language
Machine Learning
62R07
Large language models (LLMs) are powerful tools that, in a number of settings, overlap with the results of human pattern recognition and reasoning. Retrieval-augmented generation (RAG) further allows LLMs to produce tailored output depending on the contents of their RAG databases. However, LLMs depend on complex, computationally expensive algorithms. In this paper, we introduce interacting Gaussian mixture models (GMMs) as a proxy for interacting LLMs. We construct a model of interacting GMMs, complete with an analogue to RAG updating, under which GMMs can generate, exchange, and update data and parameters. We show that this interacting system of Gaussian mixture models, which can be implemented at minimal computational cost, mimics certain aspects of experimental simulations of interacting LLMs whose iterative responses depend on feedback from other LLMs. We build a Markov chain from this system of interacting GMMs; formalize and interpret the notion of polarization for such a chain; and prove lower bounds on the probability of polarization. This provides theoretical insight into the use of interacting Gaussian mixture models as a computationally efficient proxy for interacting large language models.
title Gaussian mixture models as a proxy for interacting language models
topic Computation and Language
Machine Learning
62R07
url https://arxiv.org/abs/2506.00077