mlsanityData to 14 Oct 2026

Query-aware routing for Cross-lingual performance gains in Encoders

arXiv 2610.02875 ↗First posted 2 Oct 2026cs.CL cs.AI cs.IRcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 19 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

15 of 19 references.

References · 15

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  4. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  5. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,261
  6. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,252
  7. 2024Multilingual E5 Text Embeddings: A Technical Reportcited by 372
  8. 2024jina-embeddings-v3: Multilingual Embeddings With Task LoRAcited by 141
  9. 2021Scaling Deep Contrastive Learning Batch Size under Memory Limited Setupcited by 127
  10. 2021Mr. TyDi: A Multi-lingual Benchmark for Dense Retrievalcited by 119
  11. 2022Parameter-efficient Zero-shot Transfer for Cross-Language Dense Retrieval with Adapterscited by 4
  12. 2026Improving Semantic Proximity in Information Retrieval through Cross-Lingual Alignmentcited by 3
3 more
  1. 2026CLEAR: Cross-Lingual Enhancement in Alignment via Reverse-trainingcited by 3
  2. 2024Synergistic Approach for Simultaneous Optimization of Monolingual, Cross-lingual, and Multilingual Information Retrievalcited by 2
  3. 2026Succeeding at Scale: Enterprise Retrieval Benchmark Construction and Index-Preserving Query Adaptation for Multi-Tenant Searchcited by 2

Built on by

Not cited yet.