The Scandinavian Embedding Benchmarks: Comprehensive Assessment of Multilingual and Monolingual Text Embedding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Enevoldsen, Kenneth, Kardos, Márton, Muennighoff, Niklas, Nielbo, Kristoffer Laigaard
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929373398958080
author Enevoldsen, Kenneth
Kardos, Márton
Muennighoff, Niklas
Nielbo, Kristoffer Laigaard
author_facet Enevoldsen, Kenneth
Kardos, Márton
Muennighoff, Niklas
Nielbo, Kristoffer Laigaard
contents The evaluation of English text embeddings has transitioned from evaluating a handful of datasets to broad coverage across many tasks through benchmarks such as MTEB. However, this is not the case for multilingual text embeddings due to a lack of available benchmarks. To address this problem, we introduce the Scandinavian Embedding Benchmark (SEB). SEB is a comprehensive framework that enables text embedding evaluation for Scandinavian languages across 24 tasks, 10 subtasks, and 4 task categories. Building on SEB, we evaluate more than 26 models, uncovering significant performance disparities between public and commercial solutions not previously captured by MTEB. We open-source SEB and integrate it with MTEB, thus bridging the text embedding evaluation gap for Scandinavian languages.
format Preprint
id arxiv_https___arxiv_org_abs_2406_02396
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Scandinavian Embedding Benchmarks: Comprehensive Assessment of Multilingual and Monolingual Text Embedding
Enevoldsen, Kenneth
Kardos, Márton
Muennighoff, Niklas
Nielbo, Kristoffer Laigaard
Computation and Language
Artificial Intelligence
The evaluation of English text embeddings has transitioned from evaluating a handful of datasets to broad coverage across many tasks through benchmarks such as MTEB. However, this is not the case for multilingual text embeddings due to a lack of available benchmarks. To address this problem, we introduce the Scandinavian Embedding Benchmark (SEB). SEB is a comprehensive framework that enables text embedding evaluation for Scandinavian languages across 24 tasks, 10 subtasks, and 4 task categories. Building on SEB, we evaluate more than 26 models, uncovering significant performance disparities between public and commercial solutions not previously captured by MTEB. We open-source SEB and integrate it with MTEB, thus bridging the text embedding evaluation gap for Scandinavian languages.
title The Scandinavian Embedding Benchmarks: Comprehensive Assessment of Multilingual and Monolingual Text Embedding
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.02396