# Preserving Multilingual Quality While Tuning Query Encoder on English Only

arXiv: https://arxiv.org/abs/2407.00923 · HTML: https://mlsanity.com/paper/2407.00923/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 36 references that are in the map.
- First posted on arXiv 1 Jul 2024.

## Builds on (most cited first, 24 of 24)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://mlsanity.com/paper/2212.03533/) (2022)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval](https://mlsanity.com/paper/2007.00808/) (2020)
- [Universal Sentence Encoder](https://mlsanity.com/paper/1803.11175/) (2018)
- [Large Dual Encoders Are Generalizable Retrievers](https://mlsanity.com/paper/2112.07899/) (2021)
- [Supervised Learning of Universal Sentence Representations from Natural Language Inference Data](https://mlsanity.com/paper/1705.02364/) (2017)
- [RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2010.08191/) (2020)
- [Train longer, generalize better: closing the generalization gap in large batch training of neural networks](https://mlsanity.com/paper/1705.08741/) (2017)
- [One weird trick for parallelizing convolutional neural networks](https://mlsanity.com/paper/1404.5997/) (2014)
- [Efficiently Teaching an Effective Dense Retriever with Balanced Topic Aware Sampling](https://mlsanity.com/paper/2104.06967/) (2021)
- [RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking](https://mlsanity.com/paper/2110.07367/) (2021)
- [Optimizing Dense Retrieval Model Training with Hard Negatives](https://mlsanity.com/paper/2104.08051/) (2021)
- [Choosing Transfer Languages for Cross-Lingual Learning](https://mlsanity.com/paper/1905.12688/) (2019)
- [PAIR: Leveraging Passage-Centric Similarity Relation for Improving Dense Passage Retrieval](https://mlsanity.com/paper/2108.06027/) (2021)
- [RepBERT: Contextualized Text Embeddings for First-Stage Retrieval](https://mlsanity.com/paper/2006.15498/) (2020)
- [English Contrastive Learning Can Learn Universal Cross-lingual Sentence Embeddings](https://mlsanity.com/paper/2211.06127/) (2022)
- [Empowering Dual-Encoder with Query Generator for Cross-Lingual Dense Retrieval](https://mlsanity.com/paper/2303.14991/) (2023)
- [Augmenting Passage Representations with Query Generation for Enhanced Cross-Lingual Dense Retrieval](https://mlsanity.com/paper/2305.03950/) (2023)

## Built on by (most cited first, 3 of 3)

- [LLMs are Also Effective Embedding Models: An In-depth Overview](https://mlsanity.com/paper/2412.12591/) (2024)
- [Beyond Single Embeddings: Capturing Diverse Targets with Multi-Query Retrieval](https://mlsanity.com/paper/2511.02770/) (2025)
- [SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts](https://mlsanity.com/paper/2507.13105/) (2025)

## Abstract (arXiv)

A query encoder of a dual passage retrieval system can be tuned for specific types of queries or domains, while the precomputed and stored documents representations are kept intact. Switching from one query encoder to another when needed is easily feasible, unlike overhauling the embeddings of a whole knowledge base. In this work we raise a question: Can the generic, original qualities of the encoder be preserved or at least left not too degraded when it is tuned on a narrow domain? We conducted experiments on a high quality multilingual embedding model: Tuning it on a single English-only dataset, we observe that the tuning not only preserves the multilingual qualities, but even improves them. The embedding qualities on distinctly different data are also improved or at least preserved. Drawing on our observations, we suggest a more general hypothesis: Tuning with intentionally low learning rate can preserve or improve a system's properties acquired in training, but not specifically targeted by tuning. We call this adiabatic tuning and provide tentative explanations.