Arctic-Embed 2.0: Multilingual Retrieval Without Compromise
This work addresses the challenge of maintaining high retrieval quality across languages for users in multilingual search and information retrieval applications, representing an incremental improvement over prior methods.
The paper tackles the problem of degraded English retrieval quality in multilingual embedding models by introducing Arctic-Embed 2.0, which achieves competitive retrieval performance on both multilingual and English-only benchmarks while supporting efficient storage with Matryoshka Representation Learning.
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.