Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914105583992832 |
|---|---|
| author | Wei, Yubai Han, Jiale Yang, Yi |
| author_facet | Wei, Yubai Han, Jiale Yang, Yi |
| contents | Text embedding models play a cornerstone role in AI applications, such as retrieval-augmented generation (RAG). While general-purpose text embedding models demonstrate strong performance on generic retrieval benchmarks, their effectiveness diminishes when applied to private datasets (e.g., company-specific proprietary data), which often contain specialized terminology and lingo. In this work, we introduce BMEmbed, a novel method for adapting general-purpose text embedding models to private datasets. By leveraging the well-established keyword-based retrieval technique (BM25), we construct supervisory signals from the ranking of keyword-based retrieval results to facilitate model adaptation. We evaluate BMEmbed across a range of domains, datasets, and models, showing consistent improvements in retrieval performance. Moreover, we provide empirical insights into how BM25-based signals contribute to improving embeddings by fostering alignment and uniformity, highlighting the value of this approach in adapting models to domain-specific data. We release the source code available at https://github.com/BaileyWei/BMEmbed for the research community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00363 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval Wei, Yubai Han, Jiale Yang, Yi Information Retrieval Computation and Language Text embedding models play a cornerstone role in AI applications, such as retrieval-augmented generation (RAG). While general-purpose text embedding models demonstrate strong performance on generic retrieval benchmarks, their effectiveness diminishes when applied to private datasets (e.g., company-specific proprietary data), which often contain specialized terminology and lingo. In this work, we introduce BMEmbed, a novel method for adapting general-purpose text embedding models to private datasets. By leveraging the well-established keyword-based retrieval technique (BM25), we construct supervisory signals from the ranking of keyword-based retrieval results to facilitate model adaptation. We evaluate BMEmbed across a range of domains, datasets, and models, showing consistent improvements in retrieval performance. Moreover, we provide empirical insights into how BM25-based signals contribute to improving embeddings by fostering alignment and uniformity, highlighting the value of this approach in adapting models to domain-specific data. We release the source code available at https://github.com/BaileyWei/BMEmbed for the research community. |
| title | Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval |
| topic | Information Retrieval Computation and Language |
| url | https://arxiv.org/abs/2506.00363 |