# Reconsidering SMT Over NMT for Closely Related Languages: A Case Study of Persian-Hindi Pair

arXiv: https://arxiv.org/abs/2412.16877 · HTML: https://mlsanity.com/paper/2412.16877/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 25 references that are in the map.
- First posted on arXiv 22 Dec 2024.

## Builds on (most cited first, 10 of 10)

- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [Language-agnostic BERT Sentence Embedding](https://mlsanity.com/paper/2007.01852/) (2020)
- [OpenNMT: Open-source Toolkit for Neural Machine Translation](https://mlsanity.com/paper/1709.03815/) (2017)
- [Improving Machine Translation with Phrase Pair Injection and Corpus Filtering](https://mlsanity.com/paper/2301.08008/) (2023)
- [LSCP: Enhanced Large Scale Colloquial Persian Language Understanding](https://mlsanity.com/paper/2003.06499/) (2020)
- [ParsiNorm: A Persian Toolkit for Speech Processing Normalization](https://mlsanity.com/paper/2111.03470/) (2021)
- [Learning variable length units for SMT between related languages via Byte Pair Encoding](https://mlsanity.com/paper/1610.06510/) (2016)
- [Utilizing Lexical Similarity between Related, Low-resource Languages for Pivot-based SMT](https://mlsanity.com/paper/1702.07203/) (2017)

## Abstract (arXiv)

This paper demonstrates that Phrase-Based Statistical Machine Translation (PBSMT) can outperform Transformer-based Neural Machine Translation (NMT) in moderate-resource scenarios, specifically for structurally similar languages, like the Persian-Hindi pair. Despite the Transformer architecture's typical preference for large parallel corpora, our results show that PBSMT achieves a BLEU score of 66.32, significantly exceeding the Transformer-NMT score of 53.7 on the same dataset. Additionally, we explore variations of the SMT architecture, including training on Romanized text and modifying the word order of Persian sentences to match the left-to-right (LTR) structure of Hindi. Our findings highlight the importance of choosing the right architecture based on language pair characteristics and advocate for SMT as a high-performing alternative, even in contexts commonly dominated by NMT.