About Me

I am a Research Associate at The Laboratory for Computational Social Systems (LCS2), IIIT-Delhi, advised by Prof. Tanmoy Chakraborty and Prof. Md. Shad Akhtar. I currently exploring the role of multimodality and conversational context in sarcasm detection. I am also working on NLP for code-mixed and low-resource Indian languages.

Prior to joining LCS2, I graduated with BE in Computer Engineering (2020) from PVG's College of Engineering and Technology (Afiliated to Savitribai Phule Pune University). During my undergrad, I spent my final year doing research with Prof. Manisha Marathe on Cluster Analysis of Online Mental Health Communities. I also, worked as a Research Intern (NLP and ML) at Optimum Data Analytics, Pune, working on NLP for marathi language and conversational AI for mental health chatbot.

I am keen to learn and research in the areas of Natural languge Processing and Deep Learning. I am passionate about research problems that involve understanding, interpreting, and representing unstructured data using deep learning. Recently, I am interested in self-supervised representation learning for multi-view and multi-modal data.

Please visit the Projects section to see the various projects I have worked on.

Here is my detailed CV.

Research Interests

Deep Learning, Natural Language Processing, Representation Learning, Multimodal Analysis, low-resource NLP, self-supervised learning, Computational Social Science, and Social Media Analytics.


  • PVG at WASSA 2021: A Multi-Input, Multi-Task, Transformer-Based Architecture for Empathy and Distress Prediction.
    Atharva Kulkarni, Sunanda Somwase, Shivam Rajput, and Manisha Marathe.
    In Proceedings of the EACL 2021 Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis (11th Edition of WASSA).
    [abstract] [paper] [bibtex]

  • Cluster Analysis of Online Mental Health Discourse using Topic-Infused Deep Contextualized Representations.
    Atharva Kulkarni, Amey Hengle, Pradnya Kulkarni, and Manisha Marathe.
    In Proceedings of the EACL 2021 Workshop on Health Text Mining and Information Analysis (12th Edition of LOUHI).
    [abstract] [paper] [bibtex]

  • An Attention Ensemble Approach for Efficient Text Classification of Indian Languages.
    Atharva Kulkarni, Amey Hengle, and Rutuja Udyawar.
    In Proceedings of the 17th International Conference on Natural Language Processing (ICON 2020) (TechDoFication Shared Task).
    [abstract] [paper] [bibtex]

  • Smart Cap: A Deep Learning and IoT Based Assistant for the Visually Impaired.
    Amey Hengle, Atharva Kulkarni, Nachiket Bavadekar, Niraj Kulkarni, and Rutuja Udyawar.
    In Proceedings of the Third IEEE International Conference on Smart Systems and Inventive Technology (ICSSIT 2020).
    [abstract] [paper] [bibtex]


  • Winner of the WASSA 2021 Shared Task at EACL 2021.
  • Winner of the TechDoFication Shared Task for the subtask-1f organized by ICON 2020.
  • Selected in the top 8 teams amongst 33k entries for ZS Prize Competition for the project Smart Cap.
  • Bagged the 2nd Prize in the ASPIRE 2020, a national level project competition organized by Computer Society of India (CSI) for Bachelor’s Capstone Project.


  • In my free time, I like to cook and watch cricket. I also like trekking and hiking.
  • The source code for this website was borrowed from Yuanzhe Pang.