Latest

September 2021 Paper on Is Automated Topic Model Evaluation Broken?: The Incoherence of Coherence accepted as at NeurIPS 2021 and selected for Spotlight Presentation!

June 2021 - (ongoing) Began collaborating with David Lazer and Lazer Lab on better understanding misinformation spread on Twitter.

May 2021/September 2021 Works exploring different uses of large pre-trained language models with colleagues at Microsoft Research accepted at NAACL 2021 and EMNLP 2021.

About Me

I am a budding Computer Scientist excited by research lying at the intersection of Natural Language Processing and Computational Social Science. I am a fourth-year Ph.D. student at the University of Maryland in the Computer Science program doing NLP research at the Computational Linguistics and Information Processing (CLIP) lab. I am currently working on multiple exciting projects with my advisor Philip Resnik, including text-as-data applications in computational political science, understanding misinformation narratives, and improving topic model evaluation and their ability to assist practitioners. My collaborations over these projects include the Microsoft Research, the Department of Government and Politics at UMD, and the Lazer Lab at Northeastern University. I finished my Bachelors in Computer Science and Engineering from Indian Institute of Technology (BHU) Varanasi in 2018, and completed my Master’s in CS at UMD in 2020.

Over the past year, I have been getting more and more interested in computational social science starting with applications in computational political science and understanding misinformation, with the social phenomenon of framing, including agenda-setting, a connective thread. I am (eventually) looking to delve into how to make sure NLP systems do not help create and maintain social hierarchies and systemic biases that pervade our society (much like we know language can and does) and instead look at ways NLP and computational linguistic analyses can be part of the solution (just like language can) - and acknowledging that sometimes the right thing to do is not to think about improving but scraping the technological ‘solution’ or tool.

I am lucky to have worked with many amazing people on extremely interesting topics and I look forward to exploring more!

I grew up in New Delhi, India.

Previously on ‘Latest’

September 2020 Paper on Improving Neural Topic Models using Knowledge Distillation accepted as a long paper at EMNLP 2020! Joint work with Alexander Hoyle (we are both first authors) and Philip Resnik.

August 2020 We got an NSF grant for our idea and planned project on Modeling Co-Decisions: A Computational Framework Using Language and Metadata (story).

June 2020 - August 2020 Research Internship at MSR, Redmond with Nebojsa Jojic at the NLP Group.

October 2019 Paper on Probing Pre-trained Representations for Physical Common Sense Comparisons accepted at COIN (COmmonsense INference in Natural Language Processing) at EMNLP 2019!

June 2019 - August 2019 Research Internship at MSR, Redmond with Nebojsa Jojic at the NLP Group.

August 2018 Doctoral studies (in CS) begin at University of Maryland, College Park.

May 2018 Long Paper on “How emotional are you? Neural Architectures for Emotion Intensity Prediction in Microblogs” accepted at COLING 2018!

May 2018 Gave an Oral Presentation for my paper at IWSDS 2018 on “Detecting Indirectness in Conversation”.

May 2018 Completed my Bachelors in Computer Science and Engineering at IIT (BHU) Varanasi.