SrcMix: Mixing of Related Source Languages Benefits Extremely Low-resource Machine Translation (2026.findings-eacl)
Copied to clipboard
| Challenge: | Multilingual models are widely used for machine translation, but their effectiveness for extremely low-resource languages (ELRLs) is dependent on how related languages are incorporated during fine-tuning. |
| Approach: | They propose a source-side mixing strategy that combines related ELRLs during fine-tuning while constraining the decoder to a single target language. |
| Outcome: | The proposed approach improves performance in high-resource to ELRL translations and in mid-resourced to MT translations. |
Similar Papers
CharSpan: Utilizing Lexical Similarity to Enable Zero-Shot Machine Translation for Extremely Low-resource Languages (2024.eacl-short)
Copied to clipboard
| Challenge: | Existing models for ELRLs lack parallel corpora and monolingual corporata . authors propose novel character-span noise argumentation model to facilitate cross-lingual transfer . |
| Approach: | They propose a character-span noise argumentation model to facilitate cross-lingual transfer . they use character-size noise argumentations to regularize training data of HRL . |
| Outcome: | The proposed model outperforms baselines on closely related HRL-ELRL pairs from three different language families. |
Machine Translation into Low-resource Language Varieties (2021.acl-short)
Copied to clipboard
| Challenge: | Current machine translation systems generate a "standard" target language, but many languages have multiple varieties that are different from the standard language. |
| Approach: | They propose a framework to rapidly adapt machine translation systems to generate different target varieties . they propose to use no parallel data to generate languages close to, but different from, the standard target language . |
| Outcome: | The proposed model improves on a system that generates Ukrainian and Belarusian in two languages with no parallel data. |
Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study (2021.acl-long)
Copied to clipboard
Yash Khemchandani, Sarvesh Mehtani, Vaidehi Patil, Abhijeet Awasthi, Partha Talukdar, Sunita Sarawagi
| Challenge: | Recent research in multilingual language models (LMs) has demonstrated their ability to effectively handle multiple languages in a single model. |
| Approach: | They propose to exploit relatedness among languages in a language family to overcome corpora limitations of LRLs. |
| Outcome: | The proposed model exploits relatedness among languages in a language family to overcome corpora limitations for low web-resource languages. |
Cross-Lingual Transfer Robustness to Lower-Resource Languages on Adversarial Datasets (2024.lrec-main)
Copied to clipboard
| Challenge: | Multilingual Language Models (MLLMs) exhibit robust cross-lingual transfer capabilities for downstream tasks such as Named Entity Recognition (NER) challenges persist in MLLM implementations that are not cross-linguistically robust. |
| Approach: | They evaluate two well-known MLLMs on 13 pairs of languages with a geographic, genetic, or borrowing relationship. |
| Outcome: | The proposed models show that they can leverage information acquired in a source language and apply it to a target language. |
AfriMMT-EA: Multi-domain Machine Translation for Low-Resource East African Languages (2026.findings-eacl)
Copied to clipboard
Naome A Etori, Kelechi Ezema, Nathaniel Romney Robinson, Davis David, Alfred Malengo Kondoro, Elisha Ondieki Makori, Michael Samwel Mollel, Maria Gini
| Challenge: | Recent advances in open-source large language models have demonstrated strong multilingual capabilities through data-efficient adaptation strategies. |
| Approach: | They propose to use AfriMMT-EA to refine two multilingual versions of Gemma-3 to better understand the region's linguistic and cultural diversity. |
| Outcome: | The proposed datasets comprise 54 local languages across five East African countries. |
Efficient Neural Machine Translation for Low-Resource Languages via Exploiting Related Languages (2020.acl-srw)
Copied to clipboard
| Challenge: | Neural Machine Translation (NMT) is a rapidly advancing MT paradigm that can be used to improve machine translation for many languages. |
| Approach: | They propose a technique called Unified Transliteration and Subword Segmentation to leverage language similarity while exploiting parallel data from related languages. |
| Outcome: | The proposed approach improves translation accuracy by 5 BLEU points over the standard Transformer-based NMT models. |
Efficient Low-Resource Language Adaptation via Multi-Source Dynamic Logit Fusion (2026.acl-long)
Copied to clipboard
| Challenge: | Proxy Tuning offers a logit-level strategy for introducing scaling effects, but it often fails in LRL settings because the large model’s weak LRL competence might overwhelm the knowledge of specialized smaller models. |
| Approach: | They propose a logit-based framework that balances LRL competence from a continually pretrained small model, task competence from high-resource language instruction tuning, and the scaling benefits of large models. |
| Outcome: | Experiments across four model families and eight LRLs show that TriMix outperforms single-model baselines and Proxy Tuning. |
SelectNoise: Unsupervised Noise Injection to Enable Zero-Shot Machine Translation for Extremely Low-resource Languages (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Currently, MT systems for low-resource languages lack parallel data and monolingual data. |
| Approach: | They propose an unsupervised approach to generate noisy HRLs training data by selective candidate extraction and noise injection. |
| Outcome: | The proposed model outperforms strong baselines on 12 ELRLs in a zero-shot setting . |
To Translate or Not to Translate: A Systematic Investigation of Translation-Based Cross-Lingual Transfer to Low-Resource Languages (2024.naacl-long)
Copied to clipboard
| Challenge: | XLT with multilingual language models is superfluous, says a new study . mBERT, XLM-R and mT5 are effective for cross-lingual transfer, authors say . |
| Approach: | They propose to use multilingual language models to improve cross-lingual transfer (XLT) they propose to add reliable translations to training data for XLT even for non-MT languages . |
| Outcome: | The proposed approaches outperform zero-shot XLT with mLMs, the authors show . the authors believe their findings warrant a broader inclusion of more robust translation-based baselines in XL research. |
Alignment verification to improve NMT translation towards highly inflectional languages with limited resources (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing approaches to improve translation quality using limited training data are phrase-based and syntax-based approaches. |
| Approach: | They propose to combine a neural MT system with an open source module to improve translation quality. |
| Outcome: | The proposed method improves translation quality over the best individual NMT and the standard ensemble system provided in the Marian-NMT system. |