Arctic-Embed 2.0: Multilingual Retrieval Without Compromise

TL;DR

Arctic-Embed 2.0 introduces a multilingual text embedding model that maintains high retrieval quality across languages and English, while supporting efficient storage through Matryoshka Representation Learning, addressing prior quality degradation issues.

Contribution

The paper presents Arctic-Embed 2.0's training methodology, achieving high-quality multilingual retrieval and efficient embedding storage, with open research questions and extensive experimental analysis.

Findings

Competitive retrieval quality on multilingual and English benchmarks

Supports efficient embedding storage with lower quality degradation

Addresses prior issues of degraded English retrieval quality

Abstract

This paper presents the training methodology of Arctic-Embed 2.0, a set of open-source text embedding models built for accurate and efficient multilingual retrieval. While prior works have suffered from degraded English retrieval quality, Arctic-Embed 2.0 delivers competitive retrieval quality on multilingual and English-only benchmarks, and supports Matryoshka Representation Learning (MRL) for efficient embedding storage with significantly lower compressed quality degradation compared to alternatives. We detail the design and implementation, presenting several important open research questions that arose during model development. We conduct experiments exploring these research questions and include extensive discussion aimed at fostering further discussion in this field.