"The world is full of magical things, patiently waiting for our senses to grow sharper."

About Me

Hello, I’m Tejas Vaidhya,
I am currently in my senior year of Undergraduate studies at Indian Institute of Technology, Kharagpur.

My research interests include Natural Language processing, Computer vision and Causal Inference. I am also highly interested to work on causality, explainability and Transformer based Language models. The goal of my research is to develop useful systems that work for the right reasons!

I love Programming, Designing, and Planning. I also enjoy watching Anime, reading non-fiction, film making, cooking, and interacting with new people. I am a foodie. In my free time, I like to explore food (you can often find me in some of the eateries of KGP). I have always adored technology and am very much fascinated by how it affects the life of people.

Google Scholar | Twitter Handle

News

• 7 July 2021: Attending Eastern European Machine Learning Summer School, 2021
• 1 May 2021: Mentoring two project in Google Summer of Code, 2021 and Julia Summer of Code, 2021.
• 1 March 2021: Starting Research internship at Mrinmaya’s Lab, ETH Zürich
• 14 Jan 2021: Starting internship at Uniworks Designs, Hyderabad
• 7 Jan 2021: Paper accepted to the CONSTRAINT workshop, collocated with AAAI 2021
• 6 October 2020: paper accepted to EMNLP 2020, Workshop on Noisy User-generated Text.
• 1 October 2020: paper accepted to EMNLP 2020, Workshop on Noisy User-generated Text.
• 12 September 2020: Ranked 1st at the leaderboard for EMNLP 2020, W-NUT workshop Shared Task 3.
• 31 August 2020: Completed wonderful summer at Google Summer of Code
• 31 Jun 2020: completed my research intern at CERN

Publications

• "Leveraging Event Specific and Chunk Span features to Extract COVID Events from tweets", Ayush Kaushal and Tejas Vaidhya
  Oral Presentation at the 6th Workshop on Noisy User-generated Text (W-NUT) at the 2020 Conference on Emperical Methods in Natural Language Processing (EMNLP) Shared Task WinnersMore
Twitter has acted as an important source of information during disasters and pandemic, especially during the times of COVID-19. In this paper, we describe our system entry for WNUT 2020 Shared Task-3. The task was aimed at automating the extraction of a variety of COVID-19 related events from Twitter, such as individuals who recently contracted the virus, someone with symptoms who were denied testing and believed remedies against the infection. The system consists of separate multi-task models for slot-filling subtasks and sentence-classification subtasks while leveraging the useful sentence-level information for the corresponding event. The system uses COVID-Twitter-Bert with attention-weighted pooling of candidate slot-chunk features to capture the useful information chunks. The system ranks 1st at the leader-board with F1 of 0.6598, without using any ensembles or additional datasets.

Code Pdf Cite Slides Poster


• "Domain specific BERT representation for Named Entity Recognition of lab protocol.", Tejas Vaidhya and Ayush Kaushal
  Proceedings of the 6th Workshop on Noisy User-generated Text (W-NUT) at the 2020 Conference on Emperical Methods in Natural Language Processing (EMNLP) More
Supervised models trained to predict properties from representations, have been achieving high accuracy on a variety of tasks. For instance, the BERT family seems to work exceptionally well on the downstream task from NER tagging to the range of other linguistic tasks. But the vocabulary used in the medical field contains a lot of different tokens used only in the medical industry such as the name of different diseases, devices, organisms, medicines, etc. that makes it difficult for traditional BERT model to create contextualized embedding. In this paper, we are going to illustrate the System for Named Entity Tagging based on Bio-Bert. Experimental results show that our model gives substantial improvements over the baseline and stood the fourth runner up in terms of F1 score, and first runner up in terms of Recall among 13 teams with just 2.21 F1 score behind the best one.

Code Pdf Cite Poster


• "Hostility Detection in Hindi leveraging Pre-Trained Language Models", Ojasv Kamal, Adarsh Kumar, and Tejas Vaidhya
  Oral Presentation at the Constraint workshop at Advancement Of Artificial Intelligence More
Hostile content on social platforms is ever increasing. ThisCONSTRAINT workshop at AAAI has led to the need for proper detection of hostile posts so that appropriate action can be taken to tackle them. Though a lot of work has been done recently in the English Language to solve the problem of hostile content online, similar works in Indian Languages are quite hard to find. This paper presents a transfer learning based approach to classify social media (i.e Twitter, Facebook, etc.) posts in Hindi Devanagari script as Hostile or Non-Hostile. Hostile posts are further analyzed to determine if they are Hateful, Fake, Defamation, and Offensive. This paper harnesses attention based pre-trained models fine-tuned on Hindi data with Hostile-Non hostile task as Auxiliary and fusing its features for further sub-tasks classification. Through this approach, we establish a robust and consistent model without any ensembling or complex pre-processing. We have presented the results from our approach in CONSTRAINT-2021 Shared Task[21] on hostile post detection where our model performs extremely well with 3rd runner up in terms of Weighted Fine-Grained F1 Score1.

Code Pdf Cite Slides

Experience

details of projects can be found here

Selected Projects

Kinship Relationship Prediction

Prediction of Kinship Relationship between people given the pair of images

Computer_visionKeras

Hostility and FakeNews Detection

Dectection of Fake news and Hostility in hindi leveraging pre-trained language models

NLPLanguage modelHindi

GoogleDrive.jl

Julia Package to automate Google-Drive download on top of DataDeps.jl

JuliaGoogleDrive

ALBERT Implementation in Julia

Implemented practical model of ALBERT in julia

UnderdevelopmentNLPTransformer

Energy Analysis

The assignment is aim to study the Energy analysis of residential complex for visiting reasearcher

RevitPhotoshop

Extracting COVID Entities from tweets

Proposed the system ranking 1st at the leaderboard for EMNLP 2020 W-NUT workshops shared task-3 on noisy text that utlilizes event specific and chunk span

Natural Language ProcessingDeeplearningSocial Computing

A Lite BERT and Statistical Language Models

Developed the framework for Statistical and Deeplearning Language models

NLPDeeplearningLanguage Model

NER Tagging of Lab protocols

Bio BERT based approach for NER over Wet lab protocols

Natural Language ProcessingDeeplearningLanguage Model

Identification of Taulapton

This project aim to surpass current Benchmark model of RNN.

DeeplearningNLP

Read More

You can check out the other things I do here