Saved in:
Bibliographic Details
Main Authors: Liu, Zhu, Liu, Ying, Luo, KangYang, Kong, Cunliang, Sun, Maosong
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.11380
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915292270034944
author Liu, Zhu
Liu, Ying
Luo, KangYang
Kong, Cunliang
Sun, Maosong
author_facet Liu, Zhu
Liu, Ying
Luo, KangYang
Kong, Cunliang
Sun, Maosong
contents Lexico-semantic networks represent words as nodes and their semantic relatedness as edges. While such networks are traditionally constructed using embeddings from encoder-based models or static vectors, embeddings from decoder-only large language models (LLMs) remain underexplored. Unlike encoder models, LLMs are trained with a next-token prediction objective, which does not directly encode the meaning of the current token. In this paper, we construct lexico-semantic networks from the input embeddings of LLMs with varying parameter scales and conduct a comparative analysis of their global and local structures. Our results show that these networks exhibit small-world properties, characterized by high clustering and short path lengths. Moreover, larger LLMs yield more intricate networks with less small-world effects and longer paths, reflecting richer semantic structures and relations. We further validate our approach through analyses of common conceptual pairs, structured lexical relations derived from WordNet, and a cross-lingual semantic network for qualitative words.
format Preprint
id arxiv_https___arxiv_org_abs_2502_11380
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From the New World of Word Embeddings: A Comparative Study of Small-World Lexico-Semantic Networks in LLMs
Liu, Zhu
Liu, Ying
Luo, KangYang
Kong, Cunliang
Sun, Maosong
Computation and Language
Lexico-semantic networks represent words as nodes and their semantic relatedness as edges. While such networks are traditionally constructed using embeddings from encoder-based models or static vectors, embeddings from decoder-only large language models (LLMs) remain underexplored. Unlike encoder models, LLMs are trained with a next-token prediction objective, which does not directly encode the meaning of the current token. In this paper, we construct lexico-semantic networks from the input embeddings of LLMs with varying parameter scales and conduct a comparative analysis of their global and local structures. Our results show that these networks exhibit small-world properties, characterized by high clustering and short path lengths. Moreover, larger LLMs yield more intricate networks with less small-world effects and longer paths, reflecting richer semantic structures and relations. We further validate our approach through analyses of common conceptual pairs, structured lexical relations derived from WordNet, and a cross-lingual semantic network for qualitative words.
title From the New World of Word Embeddings: A Comparative Study of Small-World Lexico-Semantic Networks in LLMs
topic Computation and Language
url https://arxiv.org/abs/2502.11380