mlsanityData to 14 Oct 2026

Rethinking what Matters: Effective and Robust Multilingual Realignment for Low-Resource Languages

arXiv 2511.06497 ↗First posted 9 Nov 2025cs.CL cs.AIcited by 0

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Realignment is a promising strategy to improve cross-lingual transfer in multilingual language models. However, empirical results are mixed and often unreliable, particularly for typologically distant or low-resource languages (LRLs) compared to English. Moreover, word realignment tools often rely on high-quality parallel data, which can be scarce or noisy for many LRLs. In this work, we conduct an extensive empirical study to investigate whether realignment truly benefits from using all available languages, or if strategically selected subsets can offer comparable or even improved cross-lingual transfer, and study the impact on LRLs. Our controlled experiments show that realignment can be particularly effective for LRLs and that using carefully selected, linguistically diverse subsets can match full multilingual alignment, and even outperform it for unseen LRLs. This indicates that effective realignment does not require exhaustive language coverage and can reduce data collection overhead, while remaining both efficient and robust when guided by informed language selection.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 25 of its 56 references that are in the map. First posted on arXiv 9 Nov 2025.

Builds on

25 of 56 references.

References · 25

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  2. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  3. 2018XNLI: Evaluating Cross-lingual Sentence Representationscited by 915
  4. 2019How multilingual is Multilingual BERT?cited by 645
  5. 2017Word Translation Without Parallel Datacited by 487
  6. 2024LLM2Vec: Large Language Models Are Secretly Powerful Text Encoderscited by 404
  7. 2019Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTcited by 301
  8. 2024A Survey on Data Selection for Language Modelscited by 236
  9. 2021Word Alignment by Fine-tuning Embeddings on Parallel Corporacited by 186
  10. 2024RegMix: Data Mixture as Regression for Language Model Pre-trainingcited by 177
  11. 2018A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddingscited by 146
  12. 2020Multilingual Alignment of Contextual Word Representationscited by 114
13 more
  1. 2018Universal Neural Machine Translation for Extremely Low Resource Languagescited by 66
  2. 2021AmericasNLI: Evaluating Zero-shot Natural Language Understanding of Pretrained Multilingual Models in Truly Low-resource Languagescited by 50
  3. 2020Inducing Language-Agnostic Multilingual Representationscited by 48
  4. 2024IrokoBench: A New Benchmark for African Languages in the Age of Large Language Modelscited by 43
  5. 2024RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMscited by 43
  6. 2021Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translationcited by 29
  7. 2025R3: Robust Rubric-Agnostic Reward Modelscited by 23
  8. 2024ProxyLM: Predicting Language Model Performance on Multilingual Tasks via Proxy Modelscited by 20
  9. 2023Exploring the Relationship between Alignment and Cross-lingual Transfer in Multilingual Transformerscited by 18
  10. 2019Target Conditioned Sampling: Optimizing Data Selection for Multilingual Neural Machine Translationcited by 11
  11. 2025Myanmar XNLI: Building a Dataset and Exploring Low-resource Approaches to Natural Language Inference with Myanmarcited by 8
  12. 2025mR3: Multilingual Rubric-Agnostic Reward Reasoning Modelscited by 8
  13. 2020Do Explicit Alignments Robustly Improve Multilingual Encoders?cited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025AlignFreeze: Navigating the Impact of Realignment on the Layers of Multilingual Models Across Diverse Languages
  2. 2024PreAlign: Boosting Cross-Lingual Transfer by Early Establishment of Multilingual Alignment
  3. 2025Middle-Layer Representation Alignment for Cross-Lingual Transfer in Fine-Tuned LLMs
  4. 2024Understanding Cross-Lingual Alignment -- A Survey
  5. 2025From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora
  6. 2026Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety Alignment
  7. 2025CM-Align: Consistency-based Multilingual Alignment for Large Language Models
  8. 2025Evaluating Multilingual and Code-Switched Alignment in LLMs via Synthetic Natural Language Inference