---
title: "LEAF — paper by Vujanic and Rückstieß (2026)"
description: "The authors present LEAF (Lightweight Embedding Alignment Framework), a distillation recipe that… 36 references, cited by 1 paper here."
authors: ["Robin Vujanic", "Thomas Rückstieß"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.2008/"
doi: "10.18653/v1/2026.acl-long.2008"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/vujanic-2026-leaf-knowledge-distillation-text-embedding/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/vujanic-2026-leaf-knowledge-distillation-text-embedding/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/) / LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations

# LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations

**Authors:** Robin Vujanic and Thomas Rückstieß · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.2008/) · [PDF](https://aclanthology.org/2026.acl-long.2008.pdf) · [DOI 10.18653/v1/2026.acl-long.2008](https://doi.org/10.18653/v1/2026.acl-long.2008) · [OpenAlex](https://openalex.org/W4415315964)

**Citations:** 1 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Text embedding models of different sizes, even within one family, usually produce incompatible vectors. Switching models therefore means re-embedding the whole collection, and retrieval systems must use the same large model for both documents and queries. The authors present LEAF (Lightweight Embedding Alignment Framework), a distillation recipe that trains a small student to reproduce its teacher's embedding space. The student is a Transformer backbone with mean pooling, a linear projection to the teacher's dimension, and normalization. Training minimizes only the ℓ2 distance between student and teacher vectors. It needs no relevance judgments, no hard negatives, no access to teacher internals and no large batches, and student and teacher may use different tokenizers or layouts. Because the vectors are compatible, documents can be indexed with the teacher while queries are encoded with the cheap student. The 23M-parameter students leaf-ir and leaf-mt rank first in their size class on BEIR and MTEB v2 (English), and asymmetric use scores higher still. This makes high-quality embeddings cheaper to train and to serve.

## Key points
- **leaf-ir** (23M) is distilled from the 109M arctic-embed-m-v1.5, a 4.7× compression. It encodes documents 6.5× faster and queries 7.3× faster, and it set a new state of the art on BEIR for models of its size.
- **leaf-mt** (23M) is distilled from the 335M mxbai-l-v1. Speed-ups are 24.4× for documents and 23.7× for queries.
  - On MTEB v2 (English), leaf-mt scores 59.4, which is 95.8% of the teacher, and leads on 5 of 7 task types.
  - In asymmetric mode it scores 60.1 (96.9% of the teacher) and leads on 6 of 7 tasks.
- Students inherit the teacher's Matryoshka (MRL) truncation behaviour and its robustness to int8 and binary quantization without being trained for either. The PCA-based alternative loses these properties.
- Each model was trained on a single A100 GPU within about 100 hours. Both are released under Apache 2.0.
- Even the best checkpoints keep a substantial validation error of about 0.3 ℓ2 on unit-norm vectors, yet downstream quality stays close to the teacher. The authors call this tolerance the system's "robustness margin."

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

Also on the byline: [MongoDB](https://ror.org/005tgre63).

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,437 works

## Cite

Cite as: Robin Vujanic and Thomas Rückstieß. 2026. LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{vujanic2026leaf,
  title = {{LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations}},
  author = {Robin Vujanic and Thomas Rückstieß},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.2008},
  url = {https://doi.org/10.18653/v1/2026.acl-long.2008}
}
```

- [← previous Law in Silico: Simulating Legal Society with LLM-Based Agents](https://md.research.elamur.ai/free/en/papers/wang-2026-law-silico-simulating-legal-society/)
- [next → Learning from Scarcity: Building and Benchmarking Speech Technology for Sukuma.](https://md.research.elamur.ai/free/en/papers/mgonzo-2026-learning-scarcity-building-benchmarking-speech/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/)

## References

36 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Daniel Campos, Alessandro Magnani, and ChengXiang Zhai. 2023. Quick dense retrievers consume kale: Post training kullback leibler alignment of embeddings for asymmetrical dual encoders. arXiv preprint arXiv:2304.01016. — [arXiv:2304.01016](https://arxiv.org/abs/2304.01016) [find](https://scholar.google.com/scholar?q=Daniel+Campos%2C+Alessandro+Magnani%2C+and+ChengXiang+Zhai%2E+2023%2E+Quick+dense+retrievers+consume+kale%3A+Post+training+kullback+leibler+alignment+of+embeddings+for+asymmetrical+dual+encoders%2E+arXiv+preprint)
2. Jianlyu Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, and Zheng Liu. 2024. [M3-embedding: Multi-linguality, multi-functionality, multi-granularity text embeddings through self-knowledge distillation](https://md.research.elamur.ai/free/en/papers/chen-2024-m3-embedding-multi-linguality/). In Findings of the Association for Computational Linguistics: ACL 2024, pages 2318–2335, Bang…
3. Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. [A simple framework for contrastive learning of visual representations](https://md.research.elamur.ai/free/en/papers/chen-2020-simple-framework-contrastive-learning-visual/). In Proceedings of the 37th International Conference on Machine Learning, ICML’20. JMLR.org.
4. Xuanang Chen, Ben He, Kai Hui, Le Sun, and Yingfei Sun. 2021. Simplified tinybert: Knowledge distillation for document retrieval. In Advances in Information Retrieval: 43rd European Conference on IR Research, ECIR 2021, Virtual Event, March 28 – April 1, 2021, Proceedings, Part II, page 241–248, Ber… [find](https://scholar.google.com/scholar?q=Xuanang+Chen%2C+Ben+He%2C+Kai+Hui%2C+Le+Sun%2C+and+Yingfei+Sun%2E+2021%2E+Simplified+tinybert%3A+Knowledge+distillation+for+document+retrieval%2E+In+Advances+in+Information+Retrieval%3A+43rd+European+Conference+on+IR+R)
5. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,…
6. Kenneth Enevoldsen, Isaac Chung, Imene Kerboua, Márton Kardos, Ashwin Mathur, David Stap, Jay Gala, Wissam Siblini, Dominik Krzemiński, Genta Indra Winata, Saba Sturua, Saiteja Utpala, Mathieu Ciancone, Marion Schaeffer, Gabriel Sequeira, Diganta Misra, Shreeya Dhakal, Jonathan Rystrøm, Roman Solom… — [MMTEB: Massive Multilingual Text Embedding Benchmark](https://md.research.elamur.ai/free/en/papers/enevoldsen-2025-mmteb-massive-multilingual-text-embedding/)
7. Mingxin Li, Yanzhao Zhang, Dingkun Long, Keqin Chen, Sibo Song, Shuai Bai, Zhibo Yang, Pengjun Xie, An Yang, Dayiheng Liu, Jingren Zhou, and Junyang Lin. 2026. Qwen3-vl-embedding and qwen3-vl-reranker: A unified framework for state-of-the-art multimodal retrieval and ranking. Preprint, arXiv:2601.04… — [arXiv:2601.04720](https://arxiv.org/abs/2601.04720) [find](https://scholar.google.com/scholar?q=Mingxin+Li%2C+Yanzhao+Zhang%2C+Dingkun+Long%2C+Keqin+Chen%2C+Sibo+Song%2C+Shuai+Bai%2C+Zhibo+Yang%2C+Pengjun+Xie%2C+An+Yang%2C+Dayiheng+Liu%2C+Jingren+Zhou%2C+and+Junyang+Lin%2E+2026%2E+Qwen3%2Dvl%2Dembedding+and+qwen3%2Dvl%2Dreranker)
8. Hugging Face. 2023. Distilbert. Accessed: 2025-05-22. [find](https://scholar.google.com/scholar?q=Hugging+Face%2E+2023%2E+Distilbert%2E+Accessed%3A+2025%2D05%2D22%2E)
9. Luke Merrick, Danmei Xu, Gaurav Nuti, and Daniel Campos. 2024. Arctic-embed: Scalable, efficient, and accurate text embedding models. arXiv preprint arXiv:2405.05374. — [arXiv:2405.05374](https://arxiv.org/abs/2405.05374) [find](https://scholar.google.com/scholar?q=Luke+Merrick%2C+Danmei+Xu%2C+Gaurav+Nuti%2C+and+Daniel+Campos%2E+2024%2E+Arctic%2Dembed%3A+Scalable%2C+efficient%2C+and+accurate+text+embedding+models%2E+arXiv+preprint+arXiv%3A2405%2E05374%2E+%E2%80%94+%5BarXiv%3A2405%2E05374%5D%28https%3A%2F%2Farxi)
10. Thibault Formal, Benjamin Piwowarski, and Stéphane Clinchant. 2021. [SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking](https://md.research.elamur.ai/free/en/papers/formal-2021-splade-sparse-lexical-expansion/), page 2288–2292. Association for Computing Machinery, New York, NY, USA.
11. Mansi Gupta, Nitish Kulkarni, Raghuveer Chanda, Anirudha Rayasam, and Zachary C Lipton. 2019. Amazonqa: A review-based question answering task. arXiv preprint arXiv:1908.04364. — [arXiv:1908.04364](https://arxiv.org/abs/1908.04364) [find](https://scholar.google.com/scholar?q=Mansi+Gupta%2C+Nitish+Kulkarni%2C+Raghuveer+Chanda%2C+Anirudha+Rayasam%2C+and+Zachary+C+Lipton%2E+2019%2E+Amazonqa%3A+A+review%2Dbased+question+answering+task%2E+arXiv+preprint+arXiv%3A1908%2E04364%2E+%E2%80%94+%5BarXiv%3A1908%2E04364%5D%28ht)
12. Sebastian Hofstätter, Sophia Althammer, Michael Schröder, Mete Sertkan, and Allan Hanbury. 2020. Improving efficient neural ranking models with cross-architecture knowledge distillation. Preprint, arXiv:2010.02666. — [arXiv:2010.02666](https://arxiv.org/abs/2010.02666) [find](https://scholar.google.com/scholar?q=Sebastian+Hofst%C3%A4tter%2C+Sophia+Althammer%2C+Michael+Schr%C3%B6der%2C+Mete+Sertkan%2C+and+Allan+Hanbury%2E+2020%2E+Improving+efficient+neural+ranking+models+with+cross%2Darchitecture+knowledge+distillation%2E+Preprint%2C+arX)
13. Hugging Face. 2024. Model distillation example for sentence transformers. [https://github.com/huggingface/sentence-transformers/](https://github.com/huggingface/sentence-transformers/) blob/main/examples/sentence_transformer/ training/distillation/model_distillation. py. Accessed: 2025-12-18. [↗](https://github.com/huggingface/sentence-transformers/)
14. Xiaoqi Jiao, Yichun Yin, Lifeng Shang, Xin Jiang, Xiao Chen, Linlin Li, Fang Wang, and Qun Liu. 2020. [TinyBERT: Distilling BERT for natural language understanding](https://md.research.elamur.ai/free/en/papers/jiao-2020-tinybert-distilling-bert-natural-language/). In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 4163– 4174, Online. Association for Computational Lingui…
15. Qiao Jin, Bhuwan Dhingra, Zhengping Liu, William Cohen, and Xinghua Lu. 2019. [Pubmedqa: A dataset for biomedical research question answering](https://md.research.elamur.ai/free/en/papers/jin-2019-pubmedqa-dataset-biomedical-research-question/). In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Proce…
16. Mandar Joshi, Eunsol Choi, Daniel Weld, and Luke Zettlemoyer. 2017. [triviaqa: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://md.research.elamur.ai/free/en/papers/joshi-2017-triviaqa-large-scale-distantly-supervised/). arXiv e-prints, arXiv:1705.03551.
17. Kaggle. 2023. Dataset containing 479k english words. [find](https://scholar.google.com/scholar?q=Kaggle%2E+2023%2E+Dataset+containing+479k+english+words%2E)
18. Aditya Kusupati, Gantavya Bhatt, Aniket Rege, Matthew Wallingford, Aditya Sinha, Vivek Ramanujan, William Howard-Snyder, Kaifeng Chen, Sham Kakade, Prateek Jain, and Ali Farhadi. 2022. Matryoshka representation learning. In Advances in Neural Information Processing Systems, volume 35, pages 30233–30… [find](https://scholar.google.com/scholar?q=Aditya+Kusupati%2C+Gantavya+Bhatt%2C+Aniket+Rege%2C+Matthew+Wallingford%2C+Aditya+Sinha%2C+Vivek+Ramanujan%2C+William+Howard%2DSnyder%2C+Kaifeng+Chen%2C+Sham+Kakade%2C+Prateek+Jain%2C+and+Ali+Farhadi%2E+2022%2E+Matryoshka+repr)
19. Niklas Muennighoff, Hongjin SU, Liang Wang, Nan Yang, Furu Wei, Tao Yu, Amanpreet Singh, and Douwe Kiela. 2025. [Generative representational instruction tuning](https://md.research.elamur.ai/free/en/papers/muennighoff-2024-generative-representational-instruction-tuning/). In The Thirteenth International Conference on Learning Representations.
20. Arvind Neelakantan, Tao Xu, Raul Puri, Alec Radford, Jesse Michael Han, Jerry Tworek, Qiming Yuan, Nikolas Tezak, Jong Wook Kim, Chris Hallacy, Johannes Heidecke, Pranav Shyam, Boris Power, Tyna Eloundou Nekoul, Girish Sastry, Gretchen Krueger, David Schnurr, Felipe Petroski Such, Kenny Hsu, and 6 o… — [Text and Code Embeddings by Contrastive Pre-Training](https://md.research.elamur.ai/free/en/papers/neelakantan-2022-text-code-embeddings-contrastive-pre/)
21. Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2016. [MS MARCO: A human generated machine reading comprehension dataset](https://md.research.elamur.ai/free/en/papers/bajaj-2016-ms-marco-human-generated/). CoRR, abs/1611.09268.
22. Jakob Nielsen. 1994. Usability engineering. Morgan Kaufmann. [find](https://scholar.google.com/scholar?q=Jakob+Nielsen%2E+1994%2E+Usability+engineering%2E+Morgan+Kaufmann%2E)
23. Guilherme Penedo, Hynek Kydlíček, Loubna Ben allal, Anton Lozhkov, Margaret Mitchell, Colin Raffel, Leandro Von Werra, and Thomas Wolf. 2024. [The fineweb datasets: Decanting the web for the finest text data at scale](https://md.research.elamur.ai/free/en/papers/penedo-2024-fineweb-datasets-decanting-web-finest/). In The Thirty-eight Conference on Neural Information Processing Systems Datasets a…
24. Nils Reimers and Iryna Gurevych. 2020. [Making monolingual sentence embeddings multilingual using knowledge distillation](https://md.research.elamur.ai/free/en/papers/reimers-2020-making-monolingual-sentence-embeddings-multilingual/). In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4512–4525, Online. Association for Computational Linguistics.
25. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2020. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. Preprint, arXiv:1910.01108. [find](https://scholar.google.com/scholar?q=Victor+Sanh%2C+Lysandre+Debut%2C+Julien+Chaumond%2C+and+Thomas+Wolf%2E+2020%2E+Distilbert%2C+a+distilled+version+of+bert%3A+smaller%2C+faster%2C+cheaper+and+lighter%2E+Preprint%2C+arXiv%3A1910%2E01108%2E)
26. Keshav Santhanam, Omar Khattab, Jon Saad-Falcon, Christopher Potts, and Matei Zaharia. 2022. [Colbertv2: Effective and efficient retrieval via lightweight late interaction](https://md.research.elamur.ai/free/en/papers/santhanam-2022-colbertv2-effective-efficient-retrieval-lightweight/). Preprint, arXiv:2112.01488.
27. Ferdinand Schlatt, Tim Hagen, Martin Potthast, and Matthias Hagen. 2025. Tite: Token-independent text encoder for information retrieval. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’25, page 2493–2503, New York, NY, USA. A… [find](https://scholar.google.com/scholar?q=Ferdinand+Schlatt%2C+Tim+Hagen%2C+Martin+Potthast%2C+and+Matthias+Hagen%2E+2025%2E+Tite%3A+Token%2Dindependent+text+encoder+for+information+retrieval%2E+In+Proceedings+of+the+48th+International+ACM+SIGIR+Conference+o)
28. Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, and Tao Yu. 2023. [One embedder, any task: Instruction-finetuned text embeddings](https://md.research.elamur.ai/free/en/papers/su-2022-one-embedder-any-task-instruction/). In Findings of the Association for Computational Linguistics: ACL 2023, pages 1102–1121, Toronto…
29. Zhiqing Sun, Hongkun Yu, Xiaodan Song, Renjie Liu, Yiming Yang, and Denny Zhou. 2020. [Mobilebert: a compact task-agnostic bert for resource-limited devices](https://md.research.elamur.ai/free/en/papers/sun-2020-mobilebert-compact-task-agnostic-bert/). arXiv preprint arXiv:2004.02984.
30. Nandan Thakur, Nils Reimers, Andreas Rücklé, Abhishek Srivastava, and Iryna Gurevych. 2021. [Beir: A heterogenous benchmark for zero-shot evaluation of information retrieval models](https://md.research.elamur.ai/free/en/papers/thakur-2021-beir-heterogenous-benchmark-zero/). arXiv preprint arXiv:2104.08663.
31. Minh-Phuc Truong, Hai An Vu, Tu Vu, Nguyen Thi Ngoc Diep, Linh Ngo Van, Thien Huu Nguyen, and Trung Le. 2025. [EMO: Embedding model distillation via intra-model relation and optimal transport alignments](https://md.research.elamur.ai/free/en/papers/truong-2025-emo-embedding-model-distillation-intra/). In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages…
32. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. [Attention is all you need](https://md.research.elamur.ai/free/en/papers/vaswani-2017-attention-all-need/). Advances in neural information processing systems, 30.
33. Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, and Ming Zhou. 2020. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. Advances in neural information processing systems, 33:5776–5788. [find](https://scholar.google.com/scholar?q=Wenhui+Wang%2C+Furu+Wei%2C+Li+Dong%2C+Hangbo+Bao%2C+Nan+Yang%2C+and+Ming+Zhou%2E+2020%2E+Minilm%3A+Deep+self%2Dattention+distillation+for+task%2Dagnostic+compression+of+pre%2Dtrained+transformers%2E+Advances+in+neural+inform)
34. Xhluca. 2023. Bm25 benchmarks. [https://github.com/xhluca/bm25-benchmarks/blob/](https://github.com/xhluca/bm25-benchmarks/blob/) 22df0ccccc083d2985a5b7e55d3ecd1764d4f9ed/ README.md?plain=1\#L220. Accessed: 2025-05-07. [↗](https://github.com/xhluca/bm25-benchmarks/blob/)
35. Jinsung Yoon and Sercan O Arik. 2025. Embedding-converter: A unified framework for cross-model embedding transformation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 25464– 25482, Vienna, Austria. Association for Computatio… [find](https://scholar.google.com/scholar?q=Jinsung+Yoon+and+Sercan+O+Arik%2E+2025%2E+Embedding%2Dconverter%3A+A+unified+framework+for+cross%2Dmodel+embedding+transformation%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Association+for+Computational+)
36. Puxuan Yu, Luke Merrick, Gaurav Nuti, and Daniel Campos. 2024. Arctic-embed 2.0: Multilingual retrieval without compromise. Preprint, arXiv:2412.04506. — [arXiv:2412.04506](https://arxiv.org/abs/2412.04506) [find](https://scholar.google.com/scholar?q=Puxuan+Yu%2C+Luke+Merrick%2C+Gaurav+Nuti%2C+and+Daniel+Campos%2E+2024%2E+Arctic%2Dembed+2%2E0%3A+Multilingual+retrieval+without+compromise%2E+Preprint%2C+arXiv%3A2412%2E04506%2E+%E2%80%94+%5BarXiv%3A2412%2E04506%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2412%2E)

*Updated 31 October 2026*