Salvato in:
Dettagli Bibliografici
Autore principale: Bradley, William F.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2411.01539
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910684961308672
author Bradley, William F.
author_facet Bradley, William F.
contents We investigate the patterns of incorrect answers produced by large language models (LLMs) during evaluation. These errors exhibit highly non-intuitive behaviors unique to each model. By analyzing these patterns, we measure the similarities between LLMs and construct a taxonomy that categorizes them based on their error correlations. Our findings reveal that the incorrect responses are not randomly distributed but systematically correlated across models, providing new insights into the underlying structures and relationships among LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2411_01539
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LLMs and the Madness of Crowds
Bradley, William F.
Computation and Language
Machine Learning
We investigate the patterns of incorrect answers produced by large language models (LLMs) during evaluation. These errors exhibit highly non-intuitive behaviors unique to each model. By analyzing these patterns, we measure the similarities between LLMs and construct a taxonomy that categorizes them based on their error correlations. Our findings reveal that the incorrect responses are not randomly distributed but systematically correlated across models, providing new insights into the underlying structures and relationships among LLMs.
title LLMs and the Madness of Crowds
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2411.01539