Abstract

Hashtag segmentation is the task of breaking a hashtag into its constituent tokens. Hashtags often encode the essence of usergenerated posts, along with information like topic and sentiment, which are useful in downstream tasks. Hashtags prioritize brevity and are written in unique ways - transliterating and mixing languages, spelling variations, creative named entities. Benchmark datasets used for the hashtag segmentation task - STAN, BOUN - are small in size and extracted from a single set of tweets. However, datasets should reflect the variations in writing styles of hashtags and also account for domain and language specificity, failing which the results will misrepresent model performance. We argue that model performance should be assessed on a wider variety of hashtags, and datasets should be carefully curated. To this end, we propose HashSet, a dataset comprising of: a) 1.9k manually annotated dataset; b) 3.3M loosely supervised dataset. HashSet dataset is sampled from a different set of tweets when compared to existing datasets and provides an alternate distribution of hashtags to build and validate hashtag segmentation models. We show that the performance of SOTA models for Hashtag Segmentation drops substantially on proposed dataset, indicating that the proposed dataset provides an alternate set of hashtags to train and assess models. Datasets and results are released publicly and can be accessed

Abstract

In this paper, we investigate the problem of text coherence modeling with a new perspective of learning better distributed sentence representation by incorporating document dependent and document independent features of a sentence. To this end, we propose a datadriven end-to-end novel neural coherence model that captures text coherence by exploiting the semantic and distributional aspects of the sentences in a document. The network is able to capture various document independent and document dependent features of a sentence which is imperative in assessing text coherence. Experiments on the standard Sentence Ordering task indicate that our proposed model shows significant performance gain of 1.7 percent in terms of accuracy score compared with the state-of-the-art baselines.

Abstract

In this paper, we describe our submissions for the Similar Language Translation Shared Task 2021. We built 3 systems in each direction for the Tamil ⇐⇒ Telugu language pair. This paper outlines experiments with various tokenization schemes to train statistical models. We also report the configuration of the submitted systems and results produced by them.

Abstract

Large pretrained models have seen enormous success in extractive summarization tasks. In this work, we investigate the influence of pretraining on a BERT-based extractive summarization system for scientific documents. We derive significant performance improvements using an intermediate pretraining step that leverages existing summarization datasets and report state-of-the-art results on a recently released scientific summarization dataset, SciTLDR. We systematically analyze the intermediate pretraining step by varying the size and domain of the pretraining corpus, changing the length of the input sequence in the target task and varying target tasks. We also investigate how intermediate pretraining interacts with contextualized word embeddings trained on different domains.

Abstract

The exponential rise of online social media has enabled the creation, distribution, and consumption of information at an unprecedented rate. However, it has also led to the burgeoning of various forms of online abuse. Increasing cases of online antisemitism have be- come one of the major concerns because of its socio-political con- sequences. Unlike other major forms of online abuse like racism, sexism, etc., online antisemitism has not been studied much from a machine learning perspective. To the best of our knowledge, we present the first work in the direction of automated multimodal de- tection of online antisemitism. The task poses multiple challenges that include extracting signals across multiple modalities, contex- tual references, and handling multiple aspects of antisemitism. Un- fortunately, there does not exist any publicly available benchmark corpus for this critical task. Hence, we collect and label two datasets with 3,102 and 3,509 social media posts from Twitter and Gab re- spectively. Further, we present a multimodal deep learning system that detects the presence of antisemitic content and its specific anti- semitism category using text and images from posts. We perform an extensive set of experiments on the two datasets to evaluate the ef- ficacy of the proposed system. Finally, we also present a qualitative analysis of our study.

Abstract

Aspect extraction is not a well-explored topic in Hindi, with only one corpus having been developed for the task. In this paper, we dis- cuss the merits of the existing corpus in terms of quality, size, sparsity, and performance in aspect extraction tasks using established mod- els. To provide a better baseline corpus for aspect extraction, we translate the SemEval 2014 aspect-based sentiment analysis dataset and annotate the aspects in that data. We provide rigorous guidelines and a replicable methodology for this task. We quantitatively evaluate the translations and annotations us- ing inter-annotator agreement scores. We also evaluate our dataset using state-of-the-art neu- ral aspect extraction models in both monolin- gual and multilingual settings and show that the models perform far better on our corpus than on the existing Hindi dataset. With this, we establish our corpus as the gold-standard aspect extraction dataset in Hindi

Abstract

In this paper, we describe our submissions for LoResMT Shared Task @MT Summit 2021 Con- ference. We built statistical translation systems in each direction for English ⇐⇒ Marathi lan- guage pair. This paper outlines initial baseline experiments with various tokenization schemes to train models. Using optimal tokenization scheme we create synthetic data and further train augmented dataset to create more statistical models. Also, we reorder English to match Marathi syntax to further train another set of baseline and data augmented models using various tok- enization schemes. We report configuration of the submitted systems and results produced by them

Abstract

Topic diversion occurs frequently with engaging open-domain dialogue systems like virtual assistants. The balance between staying on topic and rectifying the topic drift is important for a good collaborative system. In this paper, we present a model which uses a finetuned XLNet-base to classify the utterances pertaining to the major topic of conversation and those which are not, with a precision of 84%. We propose a preliminary study, classifying utterances into major, minor and offtopics, which further extends into a system initiative for diversion rectification. A case study was conducted where a system initiative is emulated as a response to the user going off-topic, mimicking a common occurrence of mixed initiative present in natural human-human conversation. This task of classifying utterances into those which belong to the major theme or not, would also help us in identification of relevant sentences for tasks like dialogue summarization and information extraction from conversations.

Abstract

Aspect extraction is not a well-explored topic in Hindi, with only one corpus having been developed for the task. In this paper, we discuss the merits of the existing corpus in terms of quality, size, sparsity, and performance in aspect extraction tasks using established models. To provide a better baseline corpus for aspect extraction, we translate the SemEval 2014 aspect-based sentiment analysis dataset and annotate the aspects in that data. We provide rigorous guidelines and a replicable methodology for this task. We quantitatively evaluate the translations and annotations using inter-annotator agreement scores. We also evaluate our dataset using state-of-the-art neural aspect extraction models in both monolingual and multilingual settings and show that the models perform far better on our corpus than on the existing Hindi dataset. With this, we establish our corpus as the gold-standard aspect extraction dataset in Hindi.

Abstract

Multiple parallel attention mechanisms that use multiple attention heads facilitate greater performance of the Transformer model for various applications e.g., Neural Machine Translation (NMT), text classification. In multi-head attention mechanism, different heads attend to different parts of the input. However, the limitation is that multiple heads might attend to the same part of the input, resulting in multiple heads being redundant. Thus, the model resources are under-utilized. One approach to avoid this is to prune least important heads based on certain importance score. In this work, we focus on designing a Dynamic Head Importance Computation Mechanism (DHICM) to dynamically calculate the importance of a head with respect to the input. Our insight is to design an additional attention layer together with multi-head attention, and utilize the outputs of the multi-head attention along with the input, to compute the importance for each head. Additionally, we add an extra loss function to prevent the model from assigning same score to all heads, to identify more important heads and improvise performance. We analyzed performance of DHICM for NMT with different languages. Experiments on different datasets show that DHICM outperforms traditional Transformer-based approach by large margin, especially, when less training data is available.