Language Models (LMs) excel in natural language processing tasks for English but show reduced performance in most other languages. This problem is commonly tackled by continually pre-training and fine-tuning these models for said languages. A significant issue in this process is the limited vocabulary coverage in the original model’s tokenizer, leading to inadequate representation of new languages and necessitating an expansion of the tokenizer. The initialization of the embeddings corresponding to new vocabulary items presents a further challenge. Current strategies require cross-lingual embeddings and lack a solid theoretical foundation as well as comparisons with strong baselines. In this paper, we first establish theoretically that initializing within the convex hull of existing embeddings is a good initialization, followed by a novel but simple approach, Constrained Word2Vec (CW2V), which does not require cross-lingual embeddings. Our study evaluates different initialization methods for expanding RoBERTa and LLaMA 2 across four languages and five tasks. The results show that CW2V performs equally well or even better than more advanced techniques. Additionally, simpler approaches like multivariate initialization perform on par with these advanced methods indicating that efficient large-scale multilingual continued pretraining can be achieved even with simpler initialization methods.

We address the task of machine translation (MT) from extremely low-resource language (ELRL) to English by leveraging cross-lingual transfer from *closely-related* high-resource language (HRL). The development of an MT system for ELRL is challenging because these languages typically lack parallel corpora and monolingual corpora, and their representations are absent from large multilingual language models. Many ELRLs share lexical similarities with some HRLs, which presents a novel modeling opportunity. However, existing subword-based neural MT models do not explicitly harness this lexical similarity, as they only implicitly align HRL and ELRL latent embedding space. To overcome this limitation, we propose a novel, CharSpan, approach based on character-span noise augmentation into the training data of HRL. This serves as a regularization technique, making the model more robust to lexical divergences between the HRL and ELRL, thus facilitating effective cross-lingual transfer. Our method significantly outperformed strong baselines in zero-shot settings on closely related HRL and ELRL pairs from three diverse language families, emerging as the state-of-the-art model for ELRLs.

This paper presents the findings of the WMT 2024’s MultiIndic22MT Shared Task, focusing on Machine Translation (MT) of 22 Indian Languages. In this task, we challenged participants with building MT systems which could translate between any or all of 22 Indian languages in the 8th schedule of the Indian constitution and English. For evaluation, we focused on automatic metrics, namely, chrF, chrF++ and BLEU.

This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages, specifically those using non-Roman scripts. We propose an approach that utilizes the romanized form of text as an interface for LLMs, hypothesizing that its frequent informal use and shared tokens with English enhance cross-lingual alignment. Our approach involve the continual pretraining of a English LLM like Llama 2 on romanized text of non-English, non-Roman script languages, followed by instruction tuning on romanized data. The results indicate that romanized text not only reduces token fertility by 2x-4x but also matches if not outperforms native script representation across various NLU, NLG and MT tasks. Moreover, the embeddings computed on romanized text exhibit closer alignment with their English translations than those from the native script. Our approach presents a promising direction for leveraging the power of English LLMs in languages traditionally underrepresented in NLP research.

Despite the considerable advancements in English LLMs, the progress in building comparable models for other languages has been hindered due to the scarcity of tailored resources. Our work aims to bridge this divide by introducing an expansive suite of resources specifically designed for the development of Indic LLMs, covering 22 languages, containing a total of 251B tokens and 74.8M instruction-response pairs. Recognizing the importance of both data quality and quantity, our approach combines highly curated manually verified data, unverified yet valuable data, and synthetic data. We build a clean, open-source pipeline for curating pre-training data from diverse sources, including websites, PDFs, and videos, incorporating best practices for crawling, cleaning, flagging, and deduplication. For instruction-fine tuning, we amalgamate existing Indic datasets, translate/transliterate English datasets into Indian languages, and utilize LLaMa2 and Mixtral models to create conversations grounded in articles from Indian Wikipedia and Wikihow. Additionally, we address toxicity alignment by generating toxic prompts for multiple scenarios and then generate non-toxic responses by feeding these toxic prompts to an aligned LLaMa2 model. We hope that the datasets, tools, and resources released as a part of this work will not only propel the research and development of Indic LLMs but also establish an open-source blueprint for extending such efforts to other languages.

While machine translation evaluation has been studied primarily for high-resource languages, there has been a recent interest in evaluation for low-resource languages due to the increasing availability of data and models. In this paper, we focus on a zero-shot evaluation setting focusing on low-resource Indian languages, namely Assamese, Kannada, Maithili, and Punjabi. We collect sufficient Multi-Dimensional Quality Metrics (MQM) and Direct Assessment (DA) annotations to create test sets and meta-evaluate a plethora of automatic evaluation metrics. We observe that even for learned metrics, which are known to exhibit zero-shot performance, the Kendall Tau and Pearson correlations with human annotations are only as high as 0.32 and 0.45. Synthetic data approaches show mixed results and overall do not help close the gap by much for these languages. This indicates that there is still a long way to go for low-resource evaluation.

The widespread online communication in a modern multilingual world has provided opportunities to blend more than one language (aka code-mixed language) in a single utterance. This has resulted a formidable challenge for the computational models due to the scarcity of annotated data and presence of noise. A potential solution to mitigate the data scarcity problem in low-resource setup is to leverage existing data in resource-rich language through translation. In this paper, we tackle the problem of code-mixed (Hinglish and Bengalish) to English machine translation. First, we synthetically develop HINMIX, a parallel corpus of Hinglish to English, with ~4.2M sentence pairs. Subsequently, we propose RCMT, a robust perturbation based joint-training model that learns to handle noise in the real-world code-mixed text by parameter sharing across clean and noisy words. Further, we show the adaptability of RCMT in a zero-shot setup for Bengalish to English translation. Our evaluation and comprehensive analyses qualitatively and quantitatively demonstrate the superiority of RCMT over state-of-the-art code-mixed and robust translation methods.

We present, Naamapadam, the largest publicly available Named Entity Recognition (NER) dataset for the 11 major Indian languages from two language families. The dataset contains more than 400k sentences annotated with a total of at least 100k entities from three standard entity categories (Person, Location, and, Organization) for 9 out of the 11 languages. The training dataset has been automatically created from the Samanantar parallel corpus by projecting automatically tagged entities from an English sentence to the corresponding Indian language translation. We also create manually annotated testsets for 9 languages. We demonstrate the utility of the obtained dataset on the Naamapadam-test dataset. We also release IndicNER, a multilingual IndicBERT model fine-tuned on Naamapadam training set. IndicNER achieves an F1 score of more than 80 for 7 out of 9 test languages. The dataset and models are available under open-source licences at https://ai4bharat.iitm.ac.in/naamapadam.

Building Natural Language Understanding (NLU) capabilities for Indic languages, which have a collective speaker base of more than one billion speakers is absolutely crucial. In this work, we aim to improve the NLU capabilities of Indic languages by making contributions along 3 important axes (i) monolingual corpora (ii) NLU testsets (iii) multilingual LLMs focusing on Indic languages. Specifically, we curate the largest monolingual corpora, IndicCorp, with 20.9B tokens covering 24 languages from 4 language families - a 2.3x increase over prior work, while supporting 12 additional languages. Next, we create a human-supervised benchmark, IndicXTREME, consisting of nine diverse NLU tasks covering 20 languages. Across languages and tasks, IndicXTREME contains a total of 105 evaluation sets, of which 52 are new contributions to the literature. To the best of our knowledge, this is the first effort towards creating a standard benchmark for Indic languages that aims to test the multilingual zero-shot capabilities of pretrained language models. Finally, we train IndicBERT v2, a state-of-the-art model supporting all the languages. Averaged across languages and tasks, the model achieves an absolute improvement of 2 points over a strong baseline. The data and models are available at https://github.com/AI4Bharat/IndicBERT.

The rapid growth of machine translation (MT) systems necessitates meta-evaluations of evaluation metrics to enable selection of those that best reflect MT quality. Unfortunately, most meta-evaluation studies focus on European languages, the observations for which may not always apply to other languages. Indian languages, having over a billion speakers, are linguistically different from them, and to date, there are no such systematic studies focused solely on English to Indian language MT. This paper fills this gap through a Multidimensional Quality Metric (MQM) dataset consisting of 7000 fine-grained annotations, spanning 5 Indian languages and 7 MT systems. We evaluate 16 metrics and show that, pre-trained metrics like COMET have the highest correlations with annotator scores as opposed to n-gram metrics like BLEU. We further leverage our MQM annotations to develop an Indic-COMET metric and show that it outperforms COMET counterparts in both human scores correlations and robustness scores in Indian languages. Additionally, we show that the Indic-COMET can outperform COMET on some unseen Indian languages. We hope that our dataset and analysis will facilitate further research in Indic MT evaluation.

We create publicly available language identification (LID) datasets and models in all 22 Indian languages listed in the Indian constitution in both native-script and romanized text. First, we create Bhasha-Abhijnaanam, a language identification test set for native-script as well as romanized text which spans all 22 Indic languages. We also train IndicLID, a language identifier for all the above-mentioned languages in both native and romanized script. For native-script text, it has better language coverage than existing LIDs and is competitive or better than other LIDs. IndicLID is the first LID for romanized text in Indian languages. Two major challenges for romanized text LID are the lack of training data and low-LID performance when languages are similar. We provide simple and effective solutions to these problems. In general, there has been limited work on romanized text in any language, and our findings are relevant to other languages that need romanized language identification. Our models are publicly available at https://github.com/AI4Bharat/IndicLID under open-source licenses. Our training and test sets are also publicly available at https://huggingface.co/datasets/ai4bharat/Bhasha-Abhijnaanam under open-source licenses.

Transliteration is very important in the Indian language context due to the usage of multiple scripts and the widespread use of romanized inputs. However, few training and evaluation sets are publicly available. We introduce Aksharantar, the largest publicly available transliteration dataset for Indian languages created by mining from monolingual and parallel corpora, as well as collecting data from human annotators. The dataset contains 26 million transliteration pairs for 21 Indic languages from 3 language families using 12 scripts. Aksharantar is 21 times larger than existing datasets and is the first publicly available dataset for 7 languages and 1 language family. We also introduce a test set of 103k word pairs for 19 languages that enables a fine-grained analysis of transliteration models on native origin words, foreign words, frequent words, and rare words. Using the training set, we trained IndicXlit, a multilingual transliteration model that improves accuracy by 15% on the Dakshina test set, and establishes strong baselines on the Aksharantar testset introduced in this work. The models, mining scripts, transliteration guidelines, and datasets are available at https://github.com/AI4Bharat/IndicXlit under open-source licenses.

Large language models have demonstrated the capability to perform on machine translation when the input is prompted with a few examples (in-context learning). Translation quality depends on various features of the selected examples, such as their quality and relevance, but previous work has predominantly focused on individual features in isolation. In this paper, we propose a general framework for combining different features influencing example selection. We learn a regression model, CTQ Scorer (Contextual Translation Quality), that selects examples based on multiple features in order to maximize the translation quality. On multiple language pairs and language models, we show that CTQ Scorer helps significantly outperform random selection as well as strong single-factor baselines reported in the literature. We also see an improvement of over 2.5 COMET points on average with respect to a strong BM25 retrieval-based baseline.

This study investigates machine translation between related languages i.e., languages within the same family that share linguistic characteristics such as word order and lexical similarity. Machine translation through few-shot prompting leverages a small set of translation pair examples to generate translations for test sentences. This procedure requires the model to learn how to generate translations while simultaneously ensuring that token ordering is maintained to produce a fluent and accurate translation. We propose that for related languages, the task of machine translation can be simplified by leveraging the monotonic alignment characteristic of such languages. We introduce DecoMT, a novel approach of few-shot prompting that decomposes the translation process into a sequence of word chunk translations. Through automatic and human evaluation conducted on multiple related language pairs across various language families, we demonstrate that our proposed approach of decomposed prompting surpasses multiple established few-shot baseline approaches. For example, DecoMT outperforms the strong few-shot prompting BLOOM model with an average improvement of 8 chrF++ scores across the examined languages.

Despite significant progress in Natural Language Generation for Indian languages (IndicNLP), there is a lack of datasets around complex structured tasks such as semantic parsing. One reason for this imminent gap is the complexity of the logical form, which makes English to multilingual translation difficult. The process involves alignment of logical forms, intents and slots with translated unstructured utterance. To address this, we propose an Inter-bilingual Seq2seq Semantic parsing dataset IE-SemParse Suite for 11 distinct Indian languages. We highlight the proposed task’s practicality, and evaluate existing multilingual seq2seq models across several train-test strategies. Our experiment reveals a high correlation across performance of original multilingual semantic parsing datasets (such as mTOP, multilingual TOP and multiATIS++) and our proposed IE-SemParse suite.

Existing research on Tabular Natural Language Inference (TNLI) exclusively examines the task in a monolingual setting where the tabular premise and hypothesis are in the same language. However, due to the uneven distribution of text resources on the web across languages, it is common to have the tabular premise in a high resource language and the hypothesis in a low resource language. As a result, we present the challenging task of bilingual Tabular Natural Language Inference (bTNLI), in which the tabular premise and a hypothesis over it are in two separate languages. We construct EI-InfoTabS: an English-Indic bTNLI dataset by translating the textual hypotheses of the English TNLI dataset InfoTabS into eleven major Indian languages. We thoroughly investigate how pre-trained multilingual models learn and perform on EI-InfoTabS. Our study shows that the performance on bTNLI can be close to its monolingual counterpart, with translate-train, translate-test and unified-train being strongly competitive baselines.

This paper presents the results of the shared tasks from the 9th workshop on Asian translation (WAT2022). For the WAT2022, 8 teams submitted their translation results for the human evaluation. We also accepted 4 research papers. About 300 translation results were submitted to the automatic evaluation server, and selected submissions were manually evaluated.

Natural Language Generation (NLG) for non-English languages is hampered by the scarcity of datasets in these languages. We present the IndicNLG Benchmark, a collection of datasets for benchmarking NLG for 11 Indic languages. We focus on five diverse tasks, namely, biography generation using Wikipedia infoboxes, news headline generation, sentence summarization, paraphrase generation and, question generation. We describe the created datasets and use them to benchmark the performance of several monolingual and multilingual baselines that leverage pre-trained sequence-to-sequence models. Our results exhibit the strong performance of multilingual language-specific pre-trained models, and the utility of models trained on our dataset for other related NLG tasks. Our dataset creation methods can be easily applied to modest-resource languages as they involve simple steps such as scraping news articles and Wikipedia infoboxes, light cleaning, and pivoting through machine translation data. To the best of our knowledge, the IndicNLG Benchmark is the first NLG benchmark for Indic languages and the most diverse multilingual NLG dataset, with approximately 8M examples across 5 tasks and 11 languages. The datasets and models will be publicly available.

While Indic NLP has made rapid advances recently in terms of the availability of corpora and pre-trained models, benchmark datasets on standard NLU tasks are limited. To this end, we introduce INDICXNLI, an NLI dataset for 11 Indic languages. It has been created by high-quality machine translation of the original English XNLI dataset and our analysis attests to the quality of INDICXNLI. By finetuning different pre-trained LMs on this INDICXNLI, we analyze various cross-lingual transfer techniques with respect to the impact of the choice of language models, languages, multi-linguality, mix-language input, etc. These experiments provide us with useful insights into the behaviour of pre-trained models for a diverse set of languages.

In this paper, we study pre-trained sequence-to-sequence models for a group of related languages, with a focus on Indic languages. We present IndicBART, a multilingual, sequence-to-sequence pre-trained model focusing on 11 Indic languages and English. IndicBART utilizes the orthographic similarity between Indic scripts to improve transfer learning between similar Indic languages. We evaluate IndicBART on two NLG tasks: Neural Machine Translation (NMT) and extreme summarization. Our experiments on NMT and extreme summarization show that a model specific to related languages like IndicBART is competitive with large pre-trained models like mBART50 despite being significantly smaller. It also performs well on very low-resource translation scenarios where languages are not included in pre-training or fine-tuning. Script sharing, multilingual training, and better utilization of limited model capacity contribute to the good performance of the compact IndicBART model.

We present Samanantar, the largest publicly available parallel corpora collection for Indic languages. The collection contains a total of 49.7 million sentence pairs between English and 11 Indic languages (from two language families). Specifically, we compile 12.4 million sentence pairs from existing, publicly available parallel corpora, and additionally mine 37.4 million sentence pairs from the Web, resulting in a 4× increase. We mine the parallel sentences from the Web by combining many corpora, tools, and methods: (a) Web-crawled monolingual corpora, (b) document OCR for extracting sentences from scanned documents, (c) multilingual representation models for aligning sentences, and (d) approximate nearest neighbor search for searching in a large collection of sentences. Human evaluation of samples from the newly mined corpora validate the high quality of the parallel sentences across 11 languages. Further, we extract 83.4 million sentence pairs between all 55 Indic language pairs from the English-centric parallel corpus using English as the pivot language. We trained multilingual NMT models spanning all these languages on Samanantar which outperform existing models and baselines on publicly available benchmarks, such as FLORES, establishing the utility of Samanantar. Our data and models are available publicly at Samanantar and we hope they will help advance research in NMT and multilingual NLP for Indic languages.

This paper presents the results of the shared tasks from the 8th workshop on Asian translation (WAT2021). For the WAT2021, 28 teams participated in the shared tasks and 24 teams submitted their translation results for the human evaluation. We also accepted 5 research papers. About 2,100 translation results were submitted to the automatic evaluation server, and selected submissions were manually evaluated.

This work introduces Itihasa, a large-scale translation dataset containing 93,000 pairs of Sanskrit shlokas and their English translations. The shlokas are extracted from two Indian epics viz., The Ramayana and The Mahabharata. We first describe the motivation behind the curation of such a dataset and follow up with empirical analysis to bring out its nuances. We then benchmark the performance of standard translation models on this corpus and show that even state-of-the-art transformer architectures perform poorly, emphasizing the complexity of the dataset.

We take up the task of large-scale evaluation of neural machine transliteration between English and Indic languages, with a focus on multilingual transliteration to utilize orthographic similarity between Indian languages. We create a corpus of 600K word pairs mined from parallel translation corpora and monolingual corpora, which is the largest transliteration corpora for Indian languages mined from public sources. We perform a detailed analysis of multilingual transliteration and propose an improved multilingual training recipe for Indic languages. We analyze various factors affecting transliteration quality like language family, transliteration direction and word origin.

The advent of neural machine translation (NMT) has opened up exciting research in building multilingual translation systems i.e. translation models that can handle more than one language pair. Many advances have been made which have enabled (1) improving translation for low-resource languages via transfer learning from high resource languages; and (2) building compact translation models spanning multiple languages. In this tutorial, we will cover the latest advances in NMT approaches that leverage multilingualism, especially to enhance low-resource translation. In particular, we will focus on the following topics: modeling parameter sharing for multi-way models, massively multilingual models, training protocols, language divergence, transfer learning, zero-shot/zero-resource learning, pivoting, multilingual pre-training and multi-source translation.

We propose a geometric framework for learning meta-embeddings of words from different embedding sources. Our framework transforms the embeddings into a common latent space, where, for example, simple averaging or concatenation of different embeddings (of a given word) is more amenable. The proposed latent space arises from two particular geometric transformations - source embedding specific orthogonal rotations and a common Mahalanobis metric scaling. Empirical results on several word similarity and word analogy benchmarks illustrate the efficacy of the proposed framework.

We describe our submission for the English→Tamil and Tamil→English news translation shared task. In this submission, we focus on exploring if a low-resource language (Tamil) can benefit from a high-resource language (Hindi) with which it shares contact relatedness. We show utilizing contact relatedness via multilingual NMT can significantly improve translation quality for English-Tamil translation.

This paper presents the results of the shared tasks from the 7th workshop on Asian translation (WAT2020). For the WAT2020, 20 teams participated in the shared tasks and 14 teams submitted their translation results for the human evaluation. We also received 12 research paper submissions out of which 7 were accepted. About 500 translation results were submitted to the automatic evaluation server, and selected submissions were manually evaluated.

In this paper, we introduce NLP resources for 11 major Indian languages from two major language families. These resources include: (a) large-scale sentence-level monolingual corpora, (b) pre-trained word embeddings, (c) pre-trained language models, and (d) multiple NLU evaluation datasets (IndicGLUE benchmark). The monolingual corpora contains a total of 8.8 billion tokens across all 11 languages and Indian English, primarily sourced from news crawls. The word embeddings are based on FastText, hence suitable for handling morphological complexity of Indian languages. The pre-trained language models are based on the compact ALBERT model. Lastly, we compile the (IndicGLUE benchmark for Indian language NLU. To this end, we create datasets for the following tasks: Article Genre Classification, Headline Prediction, Wikipedia Section-Title Prediction, Cloze-style Multiple choice QA, Winograd NLI and COPA. We also include publicly available datasets for some Indic languages for tasks like Named Entity Recognition, Cross-lingual Sentence Retrieval, Paraphrase detection, etc. Our embeddings are competitive or better than existing pre-trained embeddings on multiple tasks. We hope that the availability of the dataset will accelerate Indic NLP research which has the potential to impact more than a billion people. It can also help the community in evaluating advances in NLP over a more diverse pool of languages. The data and models are available at https://indicnlp.ai4bharat.org.

We propose a novel geometric approach for learning bilingual mappings given monolingual embeddings and a bilingual dictionary. Our approach decouples the source-to-target language transformation into (a) language-specific rotations on the original embeddings to align them in a common, latent space, and (b) a language-independent similarity metric in this common space to better model the similarity between the embeddings. Overall, we pose the bilingual mapping problem as a classification problem on smooth Riemannian manifolds. Empirically, our approach outperforms previous approaches on the bilingual lexicon induction and cross-lingual word similarity tasks. We next generalize our framework to represent multiple languages in a common latent space. Language-specific rotations for all the languages and a common similarity metric in the latent space are learned jointly from bilingual dictionaries for multiple language pairs. We illustrate the effectiveness of joint learning for multiple languages in an indirect word translation setting.

Transfer learning approaches for Neural Machine Translation (NMT) train a NMT model on an assisting language-target language pair (parent model) which is later fine-tuned for the source language-target language pair of interest (child model), with the target language being the same. In many cases, the assisting language has a different word order from the source language. We show that divergent word order adversely limits the benefits from transfer learning when little to no parallel corpus between the source and target language is available. To bridge this divergence, we propose to pre-order the assisting language sentences to match the word order of the source language and train the parent model. Our experiments on many language pairs show that bridging the word order gap leads to significant improvement in the translation quality in extremely low-resource scenarios.

This paper presents the results of the shared tasks from the 6th workshop on Asian translation (WAT2019) including Ja↔En, Ja↔Zh scientific paper translation subtasks, Ja↔En, Ja↔Ko, Ja↔En patent translation subtasks, Hi↔En, My↔En, Km↔En, Ta↔En mixed domain subtasks and Ru↔Ja news commentary translation task. For the WAT2019, 25 teams participated in the shared tasks. We also received 10 research paper submissions out of which 61 were accepted. About 400 translation results were submitted to the automatic evaluation server, and selected submis- sions were manually evaluated.

Multilingual learning for Neural Named Entity Recognition (NNER) involves jointly training a neural network for multiple languages. Typically, the goal is improving the NER performance of one of the languages (the primary language) using the other assisting languages. We show that the divergence in the tag distributions of the common named entities between the primary and assisting languages can reduce the effectiveness of multilingual learning. To alleviate this problem, we propose a metric based on symmetric KL divergence to filter out the highly divergent training instances in the assisting language. We empirically show that our data selection strategy improves NER performance in many languages, including those with very limited training data.

We address the task of joint training of transliteration models for multiple language pairs (multilingual transliteration). This is an instance of multitask learning, where individual tasks (language pairs) benefit from sharing knowledge with related tasks. We focus on transliteration involving related tasks i.e., languages sharing writing systems and phonetic properties (orthographically similar languages). We propose a modified neural encoder-decoder model that maximizes parameter sharing across language pairs in order to effectively leverage orthographic similarity. We show that multilingual transliteration significantly outperforms bilingual transliteration in different scenarios (average increase of 58% across a variety of languages we experimented with). We also show that multilingual transliteration models can generalize well to languages/language pairs not encountered during training and hence perform well on the zeroshot transliteration task. We show that further improvements can be achieved by using phonetic feature input.

We investigate pivot-based translation between related languages in a low resource, phrase-based SMT setting. We show that a subword-level pivot-based SMT model using a related pivot language is substantially better than word and morpheme-level pivot models. It is also highly competitive with the best direct translation model, which is encouraging as no direct source-target training corpus is used. We also show that combining multiple related language pivot models can rival a direct translation model. Thus, the use of subwords as translation units coupled with multiple related pivot languages can compensate for the lack of a direct parallel corpus.

We explore the use of segments learnt using Byte Pair Encoding (referred to as BPE units) as basic units for statistical machine translation between related languages and compare it with orthographic syllables, which are currently the best performing basic units for this translation task. BPE identifies the most frequent character sequences as basic units, while orthographic syllables are linguistically motivated pseudo-syllables. We show that BPE units modestly outperform orthographic syllables as units of translation, showing up to 11% increase in BLEU score. While orthographic syllables can be used only for languages whose writing systems use vowel representations, BPE is writing system independent and we show that BPE outperforms other units for non-vowel writing systems too. Our results are supported by extensive experimentation spanning multiple language families and writing systems.

In this paper, we empirically compare the two encoder-decoder neural machine translation architectures: convolutional sequence to sequence model (ConvS2S) and recurrent sequence to sequence model (RNNS2S) for English-Hindi language pair as part of IIT Bombay’s submission to WAT2017 shared task. We report the results for both English-Hindi and Hindi-English direction of language pair.

This paper describes the IIT Bombay’s submission as a part of the shared task in WAT 2016 for English–Indonesian language pair. The results reported here are for both the direction of the language pair. Among the various approaches experimented, Operation Sequence Model (OSM) and Neural Language Model have been submitted for WAT. The OSM approach integrates translation and reordering process resulting in relatively improved translation. Similarly the neural experiment integrates Neural Language Model with Statistical Machine Translation (SMT) as a feature for translation. The Neural Probabilistic Language Model (NPLM) gave relatively high BLEU points for Indonesian to English translation system while the Neural Network Joint Model (NNJM) performed better for English to Indonesian direction of translation system. The results indicate improvement over the baseline Phrase-based SMT by 0.61 BLEU points for English-Indonesian system and 0.55 BLEU points for Indonesian-English translation system.

A common and effective way to train translation systems between related languages is to consider sub-word level basic units. However, this increases the length of the sentences resulting in increased decoding time. The increase in length is also impacted by the specific choice of data format for representing the sentences as subwords. In a phrase-based SMT framework, we investigate different choices of decoder parameters as well as data format and their impact on decoding time and translation accuracy. We suggest best options for these settings that significantly improve decoding time with little impact on the translation accuracy.

We present a compendium of 110 Statistical Machine Translation systems built from parallel corpora of 11 Indian languages belonging to both Indo-Aryan and Dravidian families. We analyze the relationship between translation accuracy and the language families involved. We feel that insights obtained from this analysis will provide guidelines for creating machine translation systems of specific Indian language pairs. We build phrase based systems and some extensions. Across multiple languages, we show improvements on the baseline phrase based systems using these extensions: (1) source side reordering for English-Indian language translation, and (2) transliteration of untranslated words for Indian language-Indian language translation. These enhancements harness shared characteristics of Indian languages. To stimulate similar innovation widely in the NLP community, we have made the trained models for these language pairs publicly available.

Sufficient parallel transliteration pairs are needed for training state of the art transliteration engines. Given the cost involved, it is often infeasible to collect such data using experts. Crowdsourcing could be a cheaper alternative, provided that a good quality control (QC) mechanism can be devised for this task. Most QC mechanisms employed in crowdsourcing are aggressive (unfair to workers) and expensive (unfair to requesters). In contrast, we propose a low-cost QC mechanism which is fair to both workers and requesters. At the heart of our approach, lies a rule based Transliteration Equivalence approach which takes as input a list of vowels in the two languages and a mapping of the consonants in the two languages. We empirically show that our approach outperforms other popular QC mechanisms (viz., consensus and sampling) on two vital parameters : (i) fairness to requesters (lower cost per correct transliteration) and (ii) fairness to workers (lower rate of rejecting correct answers). Further, as an extrinsic evaluation we use the standard NEWS 2010 test set and show that such quality controlled crowdsourced data compares well to expert data when used for training a transliteration engine.

The logistics of collecting resources for Machine Translation (MT) has always been a cause of concern for some of the resource deprived languages of the world. The recent advent of crowdsourcing platforms provides an opportunity to explore the large scale generation of resources for MT. However, before venturing into this mode of resource collection, it is important to understand the various factors such as, task design, crowd motivation, quality control, etc. which can influence the success of such a crowd sourcing venture. In this paper, we present our experiences based on a series of experiments performed. This is an attempt to provide a holistic view of the different facets of translation crowd sourcing and identifying key challenges which need to be addressed for building a practical crowdsourcing solution for MT.