I am Dr. Tapas Nayak, a researcher in Natural Language Processing (NLP) and Deep Learning, currently working as a Scientist at TCS Research, Kolkata. My research focuses on enabling machines to understand, extract, and reason over information expressed in natural language.

I completed my Ph.D. in Computer Science from the National University of Singapore (NUS), where I worked with Prof. Hwee Tou Ng on developing neural models for entity and relation extraction. Following my Ph.D., I pursued a Postdoctoral Research Fellowship at IIT Kharagpur, collaborating on projects related to aspect-sentiment extraction, question-answering, causality extraction, etc.

Over the years, I have worked across a range of NLP tasks, including relation extraction, question answering, task–argument extraction, and question generation. My research has been published in top-tier conferences such as AAAI, EMNLP, CoNLL, etc. I enjoy exploring problems that combine linguistic structure, deep learning, and real-world applicability.

At TCS Research, I continue to investigate how large language models and neural architectures can enhance text understanding and knowledge extraction at scale. I also enjoy mentoring students and collaborating with academic partners on frontier NLP problems.

Outside of research, I am interested in exploring the intersections of language, cognition, and artificial intelligence, and I enjoy reading, traveling, and photography.


     

tnk02 [dot] 05 [at] gmail [dot] com


Research Experience

2021 - Present     

Research Scientist, TCS Research, India

2020 - 2021      Postdoctoral Researcher, Indian Institute of Technology Kharagpur, India
Advisor: Assoc. Prof. Pawan Goyal

Industry Experience

2012 - 2014    
Senior Consultant, PricewaterhouseCoopers India, Kolkata, India
2009 - 2012     Software Engr I, Hewlett-Packard India Software Operations, Bangalore, India

Education

2016 - 2020    

 Ph.D., Computer Science, National University of Singapore, Singapore
Thesis Title: Deep Neural Networks for Relation Extraction
Advisor: Prof. Hwee Tou Ng

2014 - 2016    

 M.E., Computer Scinece & Engineering, Jadavpur University, Kolkata, India
Thesis Title: Computer Aided Translation & Automatic Post Editing
Advisor: Assoc. Prof. Sudip Kumar Naskar
2005 - 2009     B.E., Computer Scinece & Engineering, Jadavpur University, Kolkata, India


Research Interests

  • Natural Language Understanding
  • Deep Neural Networks
  • Information Extraction
  • Aspect-Sentiment Extraction
  • Causality Extraction
  • Question-Answering
  • Neural Machine Translation
  • Knowledge Graph Completion

Publications

Few publications are listed here. For more, visit my Google Scholar page.
(* means equal contribution)
  • Effective Modeling of Encoder-Decoder Architecture for Joint Entity and Relation Extraction.
    Tapas Nayak and Hwee Tou Ng
    Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI), 2020.
    Paper Code

  • Effective Attention Modeling for Neural Relation Extraction.
    Tapas Nayak and Hwee Tou Ng
    The SIGNLL Conference on Computational Natural Language Learning (CoNLL), 2019.
    Paper Code

  • Deep Neural Approaches to Relation Triplets Extraction: A Comprehensive Survey.
    Tapas Nayak, Navonil Majumder, Pawan Goyal, and Soujanya Poria
    Cognitive Computation, 2021.
    Paper

  • Improving Distantly Supervised Relation Extraction with Self-Ensemble Noise Filtering.
    Tapas Nayak, Navonil Majumder, and Soujanya Poria
    Recent Advances in Natural Language Processing (RANLP), 2021.
    Paper Code

  • A Hierarchical Entity Graph Convolutional Network for Relation Extraction across Documents.
    Tapas Nayak and Hwee Tou Ng
    Recent Advances in Natural Language Processing (RANLP), 2021.
    Paper Code

  • Do the Benefits of Joint Models for Relation Extraction Extend to Document-level Tasks?.
    Pratik Saini, Tapas Nayak, and Indrajit Bhattacharya
    IJCNLP-AACL 2023 (short)
    Paper

  • Exploring Generative Models for Joint Attribute Value Extraction from Product Titles.
    Kalyani Roy, Tapas Nayak, and Pawan Goyal
    ArXiv 2022
    Paper

  • A Generative Approach for Financial Causality Extraction.
    Tapas Nayak, Soumya Sharma, Yash Butala, Koustuv Dasgupta, Pawan Goyal, and Niloy Ganguly
    FinWeb workshop of WWW 2022
    Paper

  • Aspect Sentiment Triplet Extraction Using Reinforcement Learning.
    Samson Yu Bai Jian, Tapas Nayak, Navonil Majumder, and Soujanya Poria
    30th ACM International Conference on Information and Knowledge Management (CIKM), 2021 (Short).
    Paper Code

  • PASTE: A Tagging-free Decoding Framework using Pointer Networks for Aspect Sentiment Triplet Extraction.
    Rajdeep Mukherjee*, Tapas Nayak*, Yash Butala, Sourangshu Bhattacharya, and Pawan Goyal
    EMNLP, 2021.
    Paper Code

  • tagE: Enabling an Embodied Agent to Understand Human Instructions.
    Chayan Sarkar, Avik Mitra, Pradip Pramanick, and Tapas Nayak
    EMNLP 2023 Findings
    Paper

  • Investigating the Generative Approach for Question Answering in E-Commerce.
    Kalyani Roy, Vineeth Kumar Balapanuru, Tapas Nayak, and Pawan Goyal
    ECNLP workshop at ACL 2022
    Paper

  • Adapting Pre-trained Generative Models for Extractive Question Answering.
    Prabir Mallick, Tapas Nayak, and Indrajit Bhattacharya
    GEM workshop @ EMNLP 2023
    Paper

  • Weakly Supervised Context-based Interview Question Generation.
    Samiran Pal, Kaamraan Khan, Avinash Kumar Singh, Subhasish Ghosh, Tapas Nayak, Girish Palshikar, and Indrajit Bhattacharya
    GEM workshop @ EMNLP 2022
    Paper

  • Unsupervised Generation of Long-form Technical Questions from Textbook Metadata using Structured Templates.
    Indrajit Bhattacharya, Subhasish Ghosh, Arpita Kundu, Pratik Saini, and Tapas Nayak
    PANDL workshop @ COLING 2022
    Paper

  • A Weak Supervision Approach for Predicting Difficulty of Technical Interview Questions.
    Arpita Kundu, Subhasish Ghosh, Pratik Saini, Tapas Nayak, and Indrajit Bhattacharya
    COLING 2022
    Paper