Metraipolis › archive › dc_e1ef6a897fbc

Query-aware routing for Cross-lingual performance gains in Encoders — arXiv:2610.02875

Source
arXiv
Original
https://arxiv.org/abs/2610.02875
License (as archived)
CC0-1.0
Archived
2026-10-05T22:50:19.142Z
SHA-256 of the archived text
45bb78098c4014da5027f05d7cbf90305e7f59488f48f6ae519c3b114bfe6f9f
Stable URL of this page
https://metraipolis.com/documento/dc_e1ef6a897fbc

How to cite

Query-aware routing for Cross-lingual performance gains in Encoders — arXiv:2610.02875 — arXiv, https://arxiv.org/abs/2610.02875 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1ef6a897fbc (Metraipolis archive dc_e1ef6a897fbc, sha256 45bb78098c4014da5027f05d7cbf90305e7f59488f48f6ae519c3b114bfe6f9f)
BibTeX
@misc{dce1ef6a897fbc,
  title = {Query-aware routing for Cross-lingual performance gains in Encoders — arXiv:2610.02875},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02875},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1ef6a897fbc; sha256 45bb78098c4014da5027f05d7cbf90305e7f59488f48f6ae519c3b114bfe6f9f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02875. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multilingual encoders can exhibit reduced retrieval effectiveness when queries and relevant documents differ in language, despite strong same-language performance. We investigate whether Finnish and Swedish cross-lingual retrieval can improve while preserving an encoder's existing same-language performance and document index. We combine a query-only low-rank adapter, trained against frozen document embeddings, with deterministic routing based on query and index languages. Cross-language queries use the adapter, while same-language queries use the original encoder. SampoTron, our fine-tuned low-rank (LoRA) adapter alongwith the Nemotron-3-Embed-1B model, improves average retrieval quality across six English, Finnish, and Swedish directions from 0.241 to 0.291 in normalized discounted cumulative gain (nDCG) at rank ten, a 20.9% relative gain on a sampled financial benchmark. All six cross-lingual directions improve, and routing preserves the original same-language performance, including two full-corpus Finnish evaluations. The approach enables selective cross-language specialization with reusable document embedding vectors.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2610.02875 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-05T22:45:55.805Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).