India has a rich linguistic landscape with languages from 4 major langua...
We create publicly available language identification (LID) datasets and
...
India is the second largest English-speaking country in the world with a...
Improving ASR systems is necessary to make new LLM-based use-cases acces...
Adapters have been positioned as a parameter-efficient fine-tuning (PEFT...
The rapid growth of machine translation (MT) systems has necessitated
co...
We present, Naamapadam, the largest publicly available Named Entity
Reco...
In this work, we introduce IndicXTREME, a benchmark consisting of nine
d...
Deep learning based text-to-speech (TTS) systems have been evolving rapi...
End-to-end (E2E) models have become the default choice for state-of-the-...
A cornerstone in AI research has been the creation and adoption of
stand...
We introduce Aksharantar, the largest publicly available transliteration...
Gesture typing is a method of typing words on a touch-based keyboard by
...
Self-attention heads are characteristic of Transformer models and have b...
In recent years, it has been seen that deep neural networks are lacking
...
Recent studies have shown the advantages of evaluating NLG systems using...
Recent methods in speech and language technology pretrain very LARGE mod...
As neural-network-based QA models become deeper and more complex, there ...
Large multilingual models, such as mBERT, have shown promise in crosslin...
Natural Language Generation (NLG) evaluation is a multifaceted task requ...
In this paper we present IndicBART, a multilingual, sequence-to-sequence...
Multilingual Language Models (MLLMs) such as mBERT, XLM, XLM-R, etc.
hav...
Multi-headed attention heads are a mainstay in transformer-based models....
Deep convolutional neural networks (CNNs) currently achieve state-of-the...
BERT and its variants have achieved state-of-the-art performance in vari...
Recent advances in Generative Adversarial Networks (GANs) have resulted ...
There is an increasing focus on model-based dialog evaluation metrics su...
The success of Deep Learning has created a surge in interest in a wide a...
The self-attention module is a key component of Transformer-based models...
Are existing object detection methods adequate for detecting text and vi...
We consider the task of generating dialogue responses from background
kn...
We present the IndicNLP corpus, a large-scale, general-domain corpus
con...
Recent studies on interpretability of attention distributions have led t...
With the prolification of multimodal interaction in various domains, rec...
Reasoning over plots by question answering (QA) is a challenging machine...
In this work, we focus on the task of Automatic Question Generation (AQG...
When humans learn to perform a difficult task (say, reading comprehensio...
The task of Reading Comprehension with Multiple Choice Questions, requir...
Recently,there has been a lot of interest in building compact models for...
Automatically evaluating the quality of dialogue responses for unstructu...
Recently there has been a lot of work on pruning filters from deep
convo...
Converting an n-dimensional vector to a probability distribution over n
...
Existing dialog datasets contain a sequence of utterances and responses
...
There has always been criticism for using n-gram based similarity metric...
There is an increasing demand for goal-oriented conversation systems whi...
Deep Learning has managed to push boundaries in a wide variety of tasks....
Semi-supervised node classification involves learning to classify unlabe...
Given a graph wherein every node has certain attributes associated with ...
Over the past few years, various tasks involving videos such as
classifi...
We propose DuoRC, a novel dataset for Reading Comprehension (RC) that
mo...