Publications

Refereed Conference/Workshop Publications (+arXiv preprint)

Semi-Structured Data (Tabular Reasoning)

Logic Consistency / Low-resource Applications

  • Logic Driven Classification for Low Resource Settings
    Shagun Uppal, Vivek Gupta, Avinash Swaminathan, Debanjan Mahata, Rakesh Gosangi, Haimin Zhang, Rajiv Ratn Shah, Amanda Stent
    Published at AACL-IJCNLP 2020 [Paper][PPT] [Video] [Data] [Code] [Media]

  • A Logic-Driven Framework for Consistency of Neural Models
    Tao Li, Vivek Gupta, Maitrey Mehta and Vivek Srikumar
    Published at EMNLP-IJCNLP 2019 [Paper] [Poster] [Code]
    (also appearing at StarAI 2020)

Fairness and Bias

  • Unbiasing Review Ratings with Tendency based Collaborative Filtering
    Pranshi Yadav*, Priya Yadav*, Pegah Nokhiz and Vivek Gupta
    Published at AACL-IJCNLP SRW 2020 [Paper] [Video] [PPT] [Code]

  • User Bias Removal in Review Score Prediction
    Rahul Wadbude, Vivek Gupta, Dheeraj Mekala, Harish Karnick
    Published at CoDS-COMAD 2018 and DAB@CIKM 2017. [Paper] [Poster] [PPT] [Code].

  • Equalizing Recourse across Groups
    Vivek Gupta*, Pegah Nokhiz*, Chitradeep Dutta Roy*, Suresh Venkatasubramanian
    Technical Report. [PrePrint]

  • Efficient Estimation of Generalization Error and Bias-Variance Components of Ensembles
    Dhruv Mahajan, Vivek Gupta, Satya Keerthi, Sundararjan Sellamanickam
    Technical Report. [PrePrint]

Long-length Document Classification

  • Improving Document Classification with Multi-Sense Embeddings,
    Vivek Gupta, Ankit Saw, Pegah Nokhiz, Harshit Gupta, and Partha Talukdar
    Published at ECAI 2020 [Paper] [Blog] [Video] [Code]
    (extention of NAACL-SRW 2019 work)

  • P-SIF: Document Embeddings using Partition Averaging
    Vivek Gupta, Ankit Saw, Pegah Nokhiz, Praneeth Netrapalli, Piyush Rai, Partha Talukdar
    Published at AAAI 2020, Presented at SustaiNLP 2020 [Paper] [Appendix] [PPT] [Poster] [Code] [Blog]

  • Word Polysemy Aware Document Vector Estimation
    Vivek Gupta, Ankit Saw, Harshit Gupta, Pegah Nokhiz and Partha Talukdar
    Presented at NAACL-SRW 2019 (non-archival)
    (extended version appear at ECAI 2020) [Code]

  • Sparse Composite Document Vectors using soft clustering over distributional representations
    Dheeraj Mekala*, Vivek Gupta*, Bhargavi Paranjape , Harish Karnick
    Published at EMNLP 2017. [Paper] [PPT] [Video] [Code]

Effective Dimentional Properties

  • On Dimensional Linguistic Properties of the Word Embedding Space
    Vikas Raunak*, Vaibhav Kumar*, Vivek Gupta and Florian Metze
    Presented at ACL-SRW 2019 (non-archival), Published at RepL4NLP 2020 [Paper] [Paper] [Code]

  • Effective Dimensionality Reduction for Word Embeddings
    Vikas Raunak, Vivek Gupta and Florian Metze
    Published at RepL4NLP 2019. [Paper] [Poster] [Code]

Summarization

eXtreme Learning (Capturing tail)

  • Distributional Semantics meet Multi-Label Learning
    Vivek Gupta, Rahul Wadbude, Nagararjan Natararjan, Harish Karnick, Prateek Jain, Piyush Rai
    Published at AAAI 2019 [Paper] [PPT] [Poster] [Extended Version] [Code]

  • Bayes-optimal Hierarichal Classification over Asymmetric Tree-Distance Loss
    Dheeraj Mekala, Vivek Gupta, Purushottam Kar, Harish Karnick
    Technical Report. [Report]

  • On Long-Tailed Phenomena in Neural Machine Translation,
    Vikas Raunak, Siddharth Dalmia, Vivek Gupta, and Florian Metze
    Published at EMNLP 2020 (findings), Presented at SPNLP2020 [Paper] [Code]

Applications

  • Product Classification in E-Commerce using Distributional Semantics
    Vivek Gupta, Harish Karnick, Ashendra Bansal, Pradhuman Jhala
    Published at COLING 2016. [Paper] [Poster] [PPT] [Code]
    Data is proprietary to company

  • Assisting Humans to Achieve Optimal Sleep by Changing Ambient Temperature
    Vivek Gupta*, Siddhant Mittal*, Sandip Bhaumik, Raj Roy
    Published at BIBM 2016, also appear in BHI 2016 and HI-DS 2016. [Paper] [PPT]
    Code and data is proprietary to company

* represent equal contribution.
some of the softwares are also reimplemented by third party for other language/dataset, contact for more information