1 code implementation • SIGDIAL (ACL) 2021 • Suvodip Dey, Maunendra Sankar Desarkar
The model parameters of Hi-DST are independent of the number of domains/slots.
no code implementations • 10 Mar 2024 • Debolena Basak, P. K. Srijith, Maunendra Sankar Desarkar
We propose TICOD, Transformer-based Image Captioning and Object detection model for jointly training both tasks by combining the losses obtained from image captioning and object detection networks.
no code implementations • 28 Jul 2023 • Kaushal Kumar Maurya, Maunendra Sankar Desarkar, Manish Gupta, Puneet Agrawal
However, such NLG models suffer from two drawbacks: (1) some of the previous session queries could be noisy and irrelevant to the user intent for the current prefix, and (2) NLG models cannot directly incorporate historical query popularity.
no code implementations • 9 May 2023 • Kaushal Kumar Maurya, Rahul Kejriwal, Maunendra Sankar Desarkar, Anoop Kunchukuttan
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).
no code implementations • 30 Dec 2022 • Arkadipta De, Satya Swaroop Gudipudi, Sourab Panchanan, Maunendra Sankar Desarkar
In this paper, we present ComplAI, a unique framework to enable, observe, analyze and quantify explainability, robustness, performance, fairness, and model behavior in drift scenarios, and to provide a single Trust Factor that evaluates different supervised Machine Learning models not just from their ability to make correct predictions but from overall responsibility perspective.
no code implementations • 12 Oct 2022 • Sharan Narasimhan, Pooja Shekar, Suvodip Dey, Maunendra Sankar Desarkar
Text Style Transfer (TST) is performable through approaches such as latent space disentanglement, cycle-consistency losses, prototype editing etc.
1 code implementation • 12 Oct 2022 • Suvodip Dey, Maunendra Sankar Desarkar, Asif Ekbal, P. K. Srijith
In this work, we propose DialoGen, a novel encoder-decoder based framework for dialogue generation with a generalized context representation that can look beyond the last-$k$ utterances.
no code implementations • 1 Oct 2022 • Manisha Dubey, P. K. Srijith, Maunendra Sankar Desarkar
We also develop a hypernetwork based continually learning temporal point process for continuous modeling of time-to-event sequences with minimal forgetting.
1 code implementation • NAACL 2022 • Sharan Narasimhan, Suvodip Dey, Maunendra Sankar Desarkar
We empirically show that this (a) produces a better organised latent space that clusters stylistically similar sentences together, (b) performs best on a diverse set of text style transfer tasks than similar denoising-inspired baselines, and (c) is capable of fine-grained control of Style Transfer strength.
1 code implementation • ACL 2022 • Suvodip Dey, Ramamohan Kummara, Maunendra Sankar Desarkar
Generally in DST, the dialogue state or belief state for a given turn contains all the intents shown by the user till that turn.
1 code implementation • 19 Mar 2022 • Kaushal Kumar Maurya, Maunendra Sankar Desarkar
In this paper, we propose a novel meta-learning framework (called Meta-X$_{NLG}$) to learn shareable structures from typologically diverse languages based on meta-learning and language clustering.
no code implementations • 6 Mar 2022 • Samujjwal Ghosh, Subhadeep Maji, Maunendra Sankar Desarkar
To overcome these challenges, we propose a multilingual disaster related text classification system which is capable to work under \{mono, cross and multi\} lingual scenarios and under limited supervision.
no code implementations • 21 Dec 2021 • Samujjwal Ghosh, Subhadeep Maji, Maunendra Sankar Desarkar
In this paper, we propose a novel way to effectively utilize labeled data from related tasks with a graph based supervised contrastive learning approach.
1 code implementation • Findings (ACL) 2021 • Kaushal Kumar Maurya, Maunendra Sankar Desarkar, Yoshinobu Kano, Kumari Deepshikha
In this framework, we further pre-train mBART sequence-to-sequence denoising auto-encoder model with an auxiliary task using monolingual data of three languages.
no code implementations • 3 Apr 2021 • Samujjwal Ghosh, Subhadeep Maji, Maunendra Sankar Desarkar
To handle this challenge, we utilize limited labeled data along with abundantly available unlabeled data, generated during a source disaster to propose a novel two-part graph neural network.
no code implementations • 15 Jan 2021 • Chander Shekhar, Bhavya Bagla, Kaushal Kumar Maurya, Maunendra Sankar Desarkar
This leads to the necessity of identifying hostile content on social media platforms.
1 code implementation • 13 Jan 2021 • Arkadipta De, Venkatesh E, Kaushal Kumar Maurya, Maunendra Sankar Desarkar
The proposed model outperformed the existing baseline models and emerged as the state-of-the-art model for detecting hostility in the Hindi posts.
1 code implementation • 13 Sep 2019 • S. VenkataKeerthy, Rohit Aggarwal, Shalini Jain, Maunendra Sankar Desarkar, Ramakrishna Upadrasta, Y. N. Srikant
As our infrastructure is based on the Intermediate Representation (IR) of the source code, obtained embeddings are both language and machine independent.
no code implementations • COLING 2018 • Sreekanth Madisetty, Maunendra Sankar Desarkar
On the other hand, most of the content in social media is non-aggressive in nature.
no code implementations • WS 2017 • Sreekanth Madisetty, Maunendra Sankar Desarkar
The final method uses lexicons, word embeddings, word n-grams, character n-grams for training the model.