Vectorizing string entries for data processing on tables: when are larger language models better? - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

Vectorizing string entries for data processing on tables: when are larger language models better?

Résumé

There are increasingly efficient data processing pipelines that work on vectors of numbers, for instance most machine learning models, or vector databases for fast similarity search. These require converting the data to numbers. While this conversion is easy for simple numerical and categorical entries, databases are strife with text entries, such as names or descriptions. In the age of large language models, what's the best strategies to vectorize tables entries, baring in mind that larger models entail more operational complexity? We study the benefits of language models in 14 analytical tasks on tables while varying the training size, as well as for a fuzzy join benchmark. We introduce a simple characterization of a column that reveals two settings: 1) a dirty categories setting, where strings share much similarities across entries, and conversely 2) a diverse entries setting. For dirty categories, pretrained language models bring little-to-no benefit compared to simpler string models. For diverse entries, we show that larger language models improve data processing. For these we investigate the complexity-performance tradeoffs and show that they reflect those of classic text embedding: larger models tend to perform better, but it is useful to fine tune them for embedding purposes.
Fichier principal
Vignette du fichier
ICDE_paper (8).pdf (595.39 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04345931 , version 1 (14-12-2023)

Identifiants

Citer

Léo Grinsztajn, Myung Jun Kim, Edouard Oyallon, Gaël Varoquaux. Vectorizing string entries for data processing on tables: when are larger language models better?. 2023. ⟨hal-04345931⟩
72 Consultations
121 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More