The Universal Dependencies (UD) project has presented itself as a valuable platform to develop various resources for the languages of the world. We present and release a sample treebank for the Indo-Aryan language of Gujarati – a widely spoken language with little linguistic resources. This treebank is the first labeled dataset for dependency parsing in the language and the script (the Gujarati script). The treebank contains 187 part-of-speech and dependency annotated sentences from diverse genres. We discuss various idiosyncratic examples, annotation choices and present an elaborate corpus along with agreement statistics. We see this work as a valuable resource and a stepping stone for research in Gujarati Computational Linguistics.

Data-to-text (D2T) generation is a natural language generation (NLG) task in which a system describes structured data in natural language. Generating natural language verbalization for structured data is challenging as the data may not contain all the required details (here, properties such as gender are missing from the input data and need to be inferred for correct language generation), and because the structured data may conflict with the knowledge contained in the LLM’s parameters learned during pre-training. Both of these factors (incorrect filling in of details, pretraining conflict and input data) can lead to so-called hallucinations. In this paper, we propose a few-shot retrieval augmented generation (RAG) system, using a symbolic retriever – PropertyRetriever. Additionally, we experiment with state-of-the-art large language models (LLMs) to generate data verbalizations. Our system achieves the best results on 4 out of 6 subtasks for METEOR and chrF++ metrics. We present our results along with an error analysis. We release our code for reproducing the results as well as the generated verbalizations from all the experiments for any further explorations here.

Data-to-text (D2T) generation describes the task of verbalizing data, often given as attribute-value pairs. While this task is relevant for many different data domains beyond the traditionally well-explored tasks of weather forecasting, restaurant recommendations, and sports reporting, a major challenge to the applicability of data-to-text generation methods is typically data sparsity. For many applications, there is extremely little training data in terms of attribute-value inputs and target language outputs available for training a model. Given the sparse data setting, recently developed prompting methods seem most suitable for addressing D2T tasks since they do not require substantial amounts of training data, unlike finetuning approaches. However, prompt-based approaches are also challenging, as a) the design and search of prompts are non-trivial; and b) hallucination problems may occur because of the strong inductive bias of these models. In this paper, we propose a retrieval-augmented modular prompt tuning () method, which constructs prompts that fit the input data closely, thereby bridging the domain gap between the large-scale language model and the structured input data. Experiments show that our method generates texts with few hallucinations and achieves state-of-the-art performance on a dataset for drone handover message generation.

We investigate the capabilities of pre-trained models, without any fine-tuning, for a document-level multilingual news similarity task of SemEval-2022. We utilize title and news content with appropriate pre-processing techniques. Our system derives 14 different similarity features using a combination of state-of-the-art methods (MPNet) with well-known statistical methods (i.e. TF-IDF, Word Mover’s distance). We formulate multilingual news similarity task as a regression task and approximate the overall similarity between two news articles using these features. Our best-performing system achieved a correlation score of 70.1% and was ranked 20th among the 34 participating teams. In this paper, in addition to a system description, we also provide further analysis of our results and an ablation study highlighting the strengths and limitations of our features. We make our code publicly available at https://github.com/cicl-iscl/multinewssimilarity

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.

Recent Deep Learning (DL) models have succeeded in achieving human-level accuracy on various natural language tasks such as question-answering, natural language inference (NLI), and textual entailment. These tasks not only require the contextual knowledge but also the reasoning abilities to be solved efficiently. In this paper, we propose an unsupervised question-answering based approach for a similar task, fact-checking. We transform the FEVER dataset into a Cloze-task by masking named entities provided in the claims. To predict the answer token, we utilize pre-trained Bidirectional Encoder Representations from Transformers (BERT). The classifier computes label based on the correctly answered questions and a threshold. Currently, the classifier is able to classify the claims as “SUPPORTS” and “MANUAL_REVIEW”. This approach achieves a label accuracy of 80.2% on the development set and 80.25% on the test set of the transformed dataset.