Research

Publications

Publications in reversed chronological order. A complete list can also be found on my Google Scholar profile.

How sensitive are translation systems to extra contexts? Mitigating gender bias in Neural Machine Translation models through relevant contexts
arXiv:2205.10762(Under Submission), 2022 Sharma, Shanya, Dey, Manan, Sinha, Koustuv

Paper

You reap what you sow: On the Challenges of Bias Evaluation Under Multilingual Settings
Challenges & Perspectives in Creating Large Language Models workshop at ACL, 2022 Talat,Zeerak, Névéol,Aurélie, Biderman,Stella, Clinciu,Miruna, Dey, Manan, Longpre,Shayne, Luccioni, Sasha, Masoud, Maraim, Mitchell, Margaret, Dragomir,Radev, Sharma, Shanya, Subramonian, Arjun, Tae, Jaesung, Tan,Samson , Tunuguntla, Deepak, and Wal, Oskar

Paper

PromptSource: An Integrated Development Environment and Repository for Natural Language Prompts
60th Annual Meeting of the Association for Computational Linguistics (ACL), Demo Track, 2022 Bach, Stephen H., Sanh, Victor, Yong, Zheng-Xin, Webson, Albert, Raffel, Colin, Nayak, Nihal V., Sharma, Abheesht, Kim, Taewoon, Bari, M Saiful, Fevry, Thibault, Alyafeai, Zaid, Dey, Manan, Santilli, Andrea, Sun, Zhiqing, Ben-David, Srulik, Xu, Canwen, Chhablani, Gunjan, Wang, Han, Fries, Jason Alan, Al-shaibani, Maged S., Sharma, Shanya, Thakker, Urmish, Almubarak, Khalid, Tang, Xiangru, Tang, Xiangru, Jiang, Mike Tian-Jian, and Rush, Alexander M

Paper

Multitask Prompted Training Enables Zero-Shot Task Generalization
International Conference on Learning Representations (Spotlight), 2022 Sanh, Victor, Webson, Albert, Raffel, Colin, Bach, Stephen, Sutawika, Lintang, Alyafeai, Zaid, Chaffin, Antoine, Stiegler, Arnaud, Raja, Arun, Dey, Manan, Bari, M Saiful, Xu, Canwen, Thakker, Urmish, Sharma, Shanya , Szczechla, Eliza, Kim, Taewoon, Chhablani, Gunjan, Nayak, Nihal, Datta, Debajyoti, Chang, Jonathan, Jiang, Mike Tian-Jian, Wang, Han, Manica, Matteo, Shen, Sheng, Yong, Zheng Xin, Pandey, Harshit, Bawden, Rachel, Wang, Thomas, Neeraj, Trishala, Rozen, Jos, Sharma, Abheesht, Santilli, Andrea, Fevry, Thibault, Fries, Jason Alan, Teehan, Ryan, Scao, Teven Le, Biderman, Stella, Gao, Leo, Wolf, Thomas, and Rush, Alexander M

Paper

Evaluating Gender Bias in Natural Language Inference
Sharma Shanya, Dey, Manan, Sinha, Koustuv
NeurIPS 2020 Workshop on Dataset Curation and Security, 2020

Paper

Assessing viewer’s mental health by detecting depression in youtube videos
Sharma Shanya, Dey, Manan
NeurIPS 2019 AISG Workshop, 2019

Paper

Building scalable mobile edge computing by enhancing quality of services
Tiwary Mayank Sharma Shanya, Mishra, Pritish
International Conference on Innovations in Information Technology (IIT), 2018

Paper

You can also check out my Github profile for a complete list of my projects.

Peer-to-peer AI-based tracing of COVID-19
Contributed to the simulator of Peer-to-Peer AI Tracing App delegated by Prof. Yoshua Bengio. The simulator simulates human mobility along with COVID-19 spreading in a city, while considering a city to consist of houses, grocery stores, parks, workplaces, and other non-essential establishments. Contributed on simulating mobility by simulating number of houses, stores, parks in a city. (Please note that my github handle, mentioned as @thechange in the repository, has been changed to @shanyas10).

Report

Code

Using Affective Computing to analyze YouTube watching history to detect depression
The project is based on the theory Misery loves company, i.e. in misery we tend to consume content such as music, literature etc. that we relate to. It is aimed at calculation of affective scores of videos using the audio visual feature and further analyzing the affective pattern generated from the YouTube watching history of an individual to predict his depression severity score.

Report

Detection of fake news during disasters
False and incorrect information during disasters can lead to chaos and panic among people on the ground and have serious detrimental outcomes for public safety.In this project, I developed a multi-modal model using BiLSTM-CHarCNN and VGG 16 that can detect the credibility of a post by effectively capturing the semantics of the text along with the features of associated piece of multimedia in it.

Report
Code

Extractive Summarizer
Developed an extraction based document summarizing application which automatically summarizes the text given by choosing the important sentences in the document.Used Naive Bayes classifierto train the model using certain statistical features- based on the frequency of some elements in text and linguistic features- based on the structure of text(length and position).

Code

Real-time facial emotion detector
(Major project for B.E. final year) Developed a facial emotion classifier using SVM to classify a facial expression as happy, sad or surprised with an accuracy of 73%. Used Viola Jones Algorithm for facial detection and Local Binary Patterns(LBP) for feature extraction. Cohn-Kanade dataset for training and evaluation purpose. To test the out-of-domain accuracy, an additional evaluation dataset of 100 images by capturing facial expressions of 5 volunteers. OOD accuracy came out to be 68%. Integrated with Arduino UNO to capture the output signal from MATLAB and display the detected emotion on 16x2 LCD

Code