BM25-Augmented Many-Shot Translation for Low-Resource North-Eastern Indian Languages

• URL: http://arxiv.org/abs/2608.13722v1
• Date: Thu, 13 Aug 2026 19:31:56 GMT
• Title: BM25-Augmented Many-Shot Translation for Low-Resource North-Eastern Indian Languages
• Abstract summary: This paper describes the University of Florida Gators submission to the WMT26 Low-Resource Indic Language Translation shared task.<n>We adapt the retrieval-augmented many-shot translation pipeline from our AmericasNLP 2026 system to translate between English and eleven North-Eastern Indian languages in both directions.
• Score: 5.3888934816887115
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This paper describes the University of Florida Gators submission to the WMT26 Low-Resource Indic Language Translation shared task. We adapt the retrieval-augmented many-shot translation pipeline from our AmericasNLP 2026 system to translate between English and eleven North-Eastern Indian languages in both directions. At inference time, BM25 retrieves the most similar parallel examples from a language-specific training bank, and Gemini 2.5 Flash translates the input conditioned on these examples. No model fine-tuning is involved. Training banks combine official WMT26 data with publicly available corpora such as Samanantar and prior WMT shared task releases. A grid search over retrieval count r and development exemplar count d across all 22 language-direction pairs selects the best configuration for each submission.

Related papers

• NLIP_Lab-IITH Multilingual MT System for WAT24 MT Shared Task [9.476463361600826]
  This paper describes NLIP Lab's multilingual machine translation system for the WAT24 shared task on multilingual Indic MT task. We explore pre-training for Indic languages using alignment agreement objectives. We fine-tuned language direction-specific multilingual translation models using small and high-quality seed data.
  arXiv  Detail & Related papers  (2024-10-17T11:18:23Z)
• NLIP_Lab-IITH Low-Resource MT System for WMT24 Indic MT Shared Task [9.476463361600826]
  We describe our system for the WMT 24 shared task of Low-Resource Indic Language Translation. Our primary system is based on language-specific finetuning on a pre-trained model. We achieve chrF2 scores of 50.6, 42.3, 54.9, and 66.3 on the official public test set for eng$rightarrow$as, eng$rightarrow$kha, eng$rightarrow$lus, eng$rightarrow$mni respectively.
  arXiv  Detail & Related papers  (2024-10-04T08:02:43Z)
• A Tulu Resource for Machine Translation [3.038642416291856]
  We present the first parallel dataset for English-Tulu translation. Tulu is spoken by approximately 2.5 million individuals in southwestern India. Our English-Tulu system, trained without using parallel English-Tulu data, outperforms Google Translate by 19 BLEU points.
  arXiv  Detail & Related papers  (2024-03-28T04:30:07Z)
• IndicTrans2: Towards High-Quality and Accessible Machine Translation Models for all 22 Scheduled Indian Languages [37.758476568195256]
  India has a rich linguistic landscape with languages from 4 major language families spoken by over a billion people. 22 of these languages are listed in the Constitution of India (referred to as scheduled languages)
  arXiv  Detail & Related papers  (2023-05-25T17:57:43Z)
• Findings of the WMT 2022 Shared Task on Translation Suggestion [63.457874930232926]
  We report the result of the first edition of the WMT shared task on Translation Suggestion. The task aims to provide alternatives for specific words or phrases given the entire documents generated by machine translation (MT) It consists two sub-tasks, namely, the naive translation suggestion and translation suggestion with hints.
  arXiv  Detail & Related papers  (2022-11-30T03:48:36Z)
• The Effect of Normalization for Bi-directional Amharic-English Neural Machine Translation [53.907805815477126]
  This paper presents the first relatively large-scale Amharic-English parallel sentence dataset. We build bi-directional Amharic-English translation models by fine-tuning the existing Facebook M2M100 pre-trained model. The results show that the normalization of Amharic homophone characters increases the performance of Amharic-English machine translation in both directions.
  arXiv  Detail & Related papers  (2022-10-27T07:18:53Z)
• A Bilingual Parallel Corpus with Discourse Annotations [82.07304301996562]
  This paper describes BWB, a large parallel corpus first introduced in Jiang et al. (2022), along with an annotated test set. The BWB corpus consists of Chinese novels translated by experts into English, and the annotated test set is designed to probe the ability of machine translation systems to model various discourse phenomena.
  arXiv  Detail & Related papers  (2022-10-26T12:33:53Z)
• MIA 2022 Shared Task: Evaluating Cross-lingual Open-Retrieval Question Answering for 16 Diverse Languages [54.002969723086075]
  We evaluate cross-lingual open-retrieval question answering systems in 16 typologically diverse languages. The best system leveraging iteratively mined diverse negative examples achieves 32.2 F1, outperforming our baseline by 4.5 points. The second best system uses entity-aware contextualized representations for document retrieval, and achieves significant improvements in Tamil (20.8 F1), whereas most of the other systems yield nearly zero scores.
  arXiv  Detail & Related papers  (2022-07-02T06:54:10Z)
• CUNI systems for WMT21: Multilingual Low-Resource Translation for Indo-European Languages Shared Task [0.0]
  We show that using joint model for multiple similar language pairs improves upon translation quality in each pair. We also demonstrate that chararacter-level bilingual models are competitive for very similar language pairs.
  arXiv  Detail & Related papers  (2021-09-20T08:10:39Z)
• Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval [51.004601358498135]
  Mr. TyDi is a benchmark dataset for mono-lingual retrieval in eleven typologically diverse languages. The goal of this resource is to spur research in dense retrieval techniques in non-English languages.
  arXiv  Detail & Related papers  (2021-08-19T16:53:43Z)
• Self-Attention with Cross-Lingual Position Representation [112.05807284056337]
  Position encoding (PE) is used to preserve the word order information for natural language processing tasks, generating fixed position indices for input sequences. Due to word order divergences in different languages, modeling the cross-lingual positional relationships might help SANs tackle this problem. We augment SANs with emphcross-lingual position representations to model the bilingually aware latent structure for the input sentence.
  arXiv  Detail & Related papers  (2020-04-28T05:23:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.