LLMs have demonstrated remarkable capability for understanding semantics, but their understanding of pragmatics is not well studied. To this end, we release a Pragmatics Understanding Benchmark (PUB) dataset consisting of fourteen tasks in four pragmatics phenomena, namely; Implicature, Presupposition, Reference, and Deixis. We curate high-quality test sets for each task, consisting of Multiple Choice Question Answers (MCQA). PUB includes a total of 28k data points, 6.1k are newly annotated. We evaluate nine models varying in the number of parameters and type of training. Our study reveals several key observations about the pragmatic capabilities of LLMs: 1. chat-fine-tuning strongly benefits smaller models, 2. large base models are competitive with their chat-fine-tuned counterparts, 3. there is a huge variance in performance across different pragmatics phenomena, and 4. a noticeable performance gap between human capabilities and model capabilities. We hope that PUB will enable comprehensive evaluation of LLM’s pragmatic reasoning capabilities.

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.

In this paper we explore the task of modeling semi-structured object sequences; in particular, we focus our attention on the problem of developing a structure-aware input representation for such sequences. Examples of such data include user activity on websites, machine logs, and many others. This type of data is often represented as a sequence of sets of key-value pairs over time and can present modeling challenges due to an ever-increasing sequence length. We propose a two-part approach, which first considers each key independently and encodes a representation of its values over time; we then self-attend over these value-aware key representations to accomplish a downstream task. This allows us to operate on longer object sequences than existing methods. We introduce a novel shared-attention-head architecture between the two modules and present an innovative training schedule that interleaves the training of both modules with shared weights for some attention heads. Our experiments on multiple prediction tasks using real-world data demonstrate that our approach outperforms a unified network with hierarchical encoding, as well as other methods including a record-centric representation and a flattened representation of the sequence.

Prompting with natural language instructions has recently emerged as a popular method of harnessing the capabilities of large language models (LLM). Given the inherent ambiguity present in natural language, it is intuitive to consider the possible advantages of prompting with less ambiguous prompt styles, like pseudo-code. In this paper, we explore if prompting via pseudo-code instructions helps improve the performance of pre-trained language models. We manually create a dataset of pseudo-code prompts for 132 different tasks spanning classification, QA, and generative language tasks, sourced from the Super-NaturalInstructions dataset. Using these prompts along with their counterparts in natural language, we study their performance on two LLM families - BLOOM, CodeGen. Our experiments show that using pseudo-code instructions leads to better results, with an average increase (absolute) of 7-16 points in F1 scores for classification tasks and an improvement (relative) of 12-38% in aggregate ROUGE-L scores across all tasks. We include detailed ablation studies which indicate that code comments, docstrings, and the structural clues encoded in pseudo-code all contribute towards the improvement in performance. To the best of our knowledge, our work is the first to demonstrate how pseudo-code prompts can be helpful in improving the performance of pre-trained LMs.

The proliferation of fake news poses a significant challenge in the digital era. Detecting false information, especially in non-English languages, is crucial to combating misinformation effectively. In this research, we introduce a novel approach for Dravidian fake news detection by harnessing the capabilities of the MuRIL transformer model, further enhanced by gradient accumulation techniques. Our study focuses on the Dravidian languages, a diverse group of languages spoken in South India, which are often underserved in natural language processing research. We optimize memory usage, stabilize training, and improve the model’s overall performance by accumulating gradients over multiple batches. The proposed model exhibits promising results in terms of both accuracy and efficiency. Our findings underline the significance of adapting state-of-the-art techniques, such as MuRIL-based models and gradient accumulation, to non-English language.

Recent years have seen a proliferation of aggressive social media posts, often wreaking even real-world consequences for victims. Aggressive behaviour on social media is especially evident during important sociopolitical events such as elections, communal incidents, and public protests. In this paper, we introduce a dataset in English to model political aggression. The dataset comprises public tweets collated across the time-frames of two of the most recent Indian general elections. We manually annotate this data for the task of aggression detection and analyze this data for aggressive behaviour. To benchmark the efficacy of our dataset, we perform experiments by fine-tuning pre-trained language models and comparing the results with models trained on an existing but general domain dataset. Our models consistently outperform the models trained on existing data. Our best model achieves a macro F1-score of 66.66 on our dataset. We also train models on a combined version of both datasets, achieving best macro F1-score of 92.77, on our dataset. Additionally, we create subsets of code-mixed and non-code-mixed data from the combined dataset to observe variations in results due to the Hindi-English code-mixing phenomenon. We publicly release the anonymized data, code, and models for further research.

In this paper, we compare two approaches to train a multilingual language model: (i) simple multilingual learning using data-mixing, and (ii) meta-learning. We examine the performance of these models by extending them to unseen language pairs and further finetune them for the task of unsupervised NMT. We perform several experiments with varying amounts of data and give a comparative analysis of the approaches. We observe that both approaches give a comparable performance, and meta-learning gives slightly better results in a few cases of low amounts of data. For Oriya-Punjabi language pair, meta-learning performs better than multilingual learning when using 2M, and 3M sentences.

Cyberbullying is a serious societal issue widespread on various channels and platforms, particularly social networking sites. Such platforms have proven to be exceptionally fertile grounds for such behavior. The dearth of high-quality training data for multilingual and low-resource scenarios, data that can accurately capture the nuances of social media conversations, often poses a roadblock to this task. This paper attempts to tackle cyberbullying, specifically its two most common manifestations - aggression and offensiveness. We present a novel, manually annotated dataset of a total of 10,000 English and Hindi-English code-mixed tweets, manually annotated for aggression detection and offensive language detection tasks. Our annotations are supported by inter-annotator agreement scores of 0.67 and 0.74 for the two tasks, indicating substantial agreement. We perform comprehensive fine-tuning of pre-trained language models (PTLMs) using this dataset to check its efficacy. Our challenging test sets show that the best models achieve macro F1-scores of 67.87 and 65.45 on the two tasks, respectively. Further, we perform cross-dataset transfer learning to benchmark our dataset against existing aggression and offensive language datasets. We also present a detailed quantitative and qualitative analysis of errors in prediction, and with this paper, we publicly release the novel dataset, code, and models.

Performance of downstream NLP tasks on code-switched Hindi-English (aka ) continues to remain a significant challenge. Intuitively, Hindi and English corpora should aid improve task performance on Hinglish. We show that meta-learning framework can effectively utilize the the labelled resources of the downstream tasks in the constituent languages. The proposed approach improves the performance on downstream tasks on code-switched language. We experiment with code-switching benchmark GLUECoS and report significant improvements.

Named Entity Recognition (NER) is a foundational NLP task that aims to provide class labels like Person, Location, Organisation, Time, and Number to words in free text. Named Entities can also be multi-word expressions where the additional I-O-B annotation information helps label them during the NER annotation process. While English and European languages have considerable annotated data for the NER task, Indian languages lack on that front- both in terms of quantity and following annotation standards. This paper releases a significantly sized standard-abiding Hindi NER dataset containing 109,146 sentences and 2,220,856 tokens, annotated with 11 tags. We discuss the dataset statistics in all their essential detail and provide an in-depth analysis of the NER tag-set used with our data. The statistics of tag-set in our dataset shows a healthy per-tag distribution especially for prominent classes like Person, Location and Organisation. Since the proof of resource-effectiveness is in building models with the resource and testing the model on benchmark data and against the leader-board entries in shared tasks, we do the same with the aforesaid data. We use different language models to perform the sequence labelling task for NER and show the efficacy of our data by performing a comparative evaluation with models trained on another dataset available for the Hindi NER task. Our dataset helps achieve a weighted F1 score of 88.78 with all the tags and 92.22 when we collapse the tag-set, as discussed in the paper. To the best of our knowledge, no available dataset meets the standards of volume (amount) and variability (diversity), as far as Hindi NER is concerned. We fill this gap through this work, which we hope will significantly help NLP for Hindi. We release this dataset with our code and models for further research at https://github.com/cfiltnlp/HiNER

This paper describes our submission for the shared task on Unsupervised MT and Very Low Resource Supervised MT at WMT 2021. We submitted systems for two language pairs: German ↔ Upper Sorbian (de ↔ hsb) and German-Lower Sorbian (de ↔ dsb). For de ↔ hsb, we pretrain our system using MASS (Masked Sequence to Sequence) objective and then finetune using iterative back-translation. Final finetunng is performed using the parallel data provided for translation objective. For de ↔ dsb, no parallel data is provided in the task, we use final de ↔ hsb model as initialization of the de ↔ dsb model and train it further using iterative back-translation, using the same vocabulary as used in the de ↔ hsb model.

We explore the impact of leveraging the relatedness of languages that belong to the same family in NLP models using multilingual fine-tuning. We hypothesize and validate that multilingual fine-tuning of pre-trained language models can yield better performance on downstream NLP applications, compared to models fine-tuned on individual languages. A first of its kind detailed study is presented to track performance change as languages are added to a base language in a graded and greedy (in the sense of best boost of performance) manner; which reveals that careful selection of subset of related languages can significantly improve performance than utilizing all related languages. The Indo-Aryan (IA) language family is chosen for the study, the exact languages being Bengali, Gujarati, Hindi, Marathi, Oriya, Punjabi and Urdu. The script barrier is crossed by simple rule-based transliteration of the text of all languages to Devanagari. Experiments are performed on mBERT, IndicBERT, MuRIL and two RoBERTa-based LMs, the last two being pre-trained by us. Low resource languages, such as Oriya and Punjabi, are found to be the largest beneficiaries of multilingual fine-tuning. Textual Entailment, Entity Classification, Section Title Prediction, tasks of IndicGLUE and POS tagging form our test bed. Compared to monolingual fine tuning we get relative performance improvement of up to 150% in the downstream tasks. The surprise take-away is that for any language there is a particular combination of other languages which yields the best performance, and any additional language is in fact detrimental.

The gaze behaviour of a reader is helpful in solving several NLP tasks such as automatic essay grading. However, collecting gaze behaviour from readers is costly in terms of time and money. In this paper, we propose a way to improve automatic essay grading using gaze behaviour, which is learnt at run time using a multi-task learning framework. To demonstrate the efficacy of this multi-task learning based approach to automatic essay grading, we collect gaze behaviour for 48 essays across 4 essay sets, and learn gaze behaviour for the rest of the essays, numbering over 7000 essays. Using the learnt gaze behaviour, we can achieve a statistically significant improvement in performance over the state-of-the-art system for the essay sets where we have gaze data. We also achieve a statistically significant improvement for 4 other essay sets, numbering about 6000 essays, where we have no gaze behaviour data available. Our approach establishes that learning gaze behaviour improves automatic essay grading.

Automatic essay grading (AEG) is a process in which machines assign a grade to an essay written in response to a topic, called the prompt. Zero-shot AEG is when we train a system to grade essays written to a new prompt which was not present in our training data. In this paper, we describe a solution to the problem of zero-shot automatic essay grading, using cognitive information, in the form of gaze behaviour. Our experiments show that using gaze behaviour helps in improving the performance of AEG systems, especially when we provide a new essay written in response to a new prompt for scoring, by an average of almost 5 percentage points of QWK.

A noun compound is a sequence of contiguous nouns that acts as a single noun, although the predicate denoting the semantic relation between its components is dropped. Noun Compound Interpretation is the task of uncovering the relation, in the form of a preposition or a free paraphrase. Prepositional paraphrasing refers to the use of preposition to explain the semantic relation, whereas free paraphrasing refers to invoking an appropriate predicate denoting the semantic relation. In this paper, we propose an unsupervised methodology for these two types of paraphrasing. We use pre-trained contextualized language models to uncover the ‘missing’ words (preposition or predicate). These language models are usually trained to uncover the missing word/words in a given input sentence. Our approach uses templates to prepare the input sequence for the language model. The template uses a special token to indicate the missing predicate. As the model has already been pre-trained to uncover a missing word (or a sequence of words), we exploit it to predict missing words for the input sequence. Our experiments using four datasets show that our unsupervised approach (a) performs comparably to supervised approaches for prepositional paraphrasing, and (b) outperforms supervised approaches for free paraphrasing. Paraphrasing (prepositional or free) using our unsupervised approach is potentially helpful for NLP tasks like machine translation and information extraction.

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.

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.