Aakanksha Naik

Carnegie Mellon University · Pittsburgh, PA 15217 · anaik@cs.cmu.edu

I am a PhD student in the Language Technologies Institute at the School of Computer Science in Carnegie Mellon University. My research interests lie in the area of natural language understanding, especially event extraction and ordering, summarization, dialogue systems and natural language inference. I am lucky to be advised by Prof Carolyn Rose. In the past, I have also worked with Prof Ed Hovy, Prof Graham Neubig, Prof Eric Nyberg, Prof Alan Black and Prof James Antaki on various projects.

Prior to starting my PhD, I received my Masters in Language Technologies from Carnegie Mellon University in August 2018. Before joining CMU, I received my BE (bachelor of engineering) in Computer Science from Birla Institute of Technology and Science, Pilani in August 2016. During my undergrad, I had the opportunity to intern with the Machine and Language Learning (MALL) Lab at the Indian Institute of Science (IISc) and the Natural Language Dialog group at the Institute for Creative Technologies, University of Southern California.

News:

  • Our work on developing a benchmark evaluation platform for numerical reasoning in natural language inference was accepted to CoNLL 2019!
  • Our work on building a dataset for discourse-level temporal ordering of events will be presented at SIGDIAL 2019. See you in Sweden!
  • Completed my internship with the Assistant Reasoning group at Facebook!

    Publications

  1. EQUATE: A Benchmark Evaluation Platform for Numerical Reasoning in Natural Language Inference
    Abhilasha Ravichander*, Aakanksha Naik*, Carolyn Rose, Eduard Hovy
    [pdf, code/data]
  2. CoNLL 2019
  3. TDDiscourse: A Dataset for Discourse-Level Temporal Ordering of Events
    Aakanksha Naik, Luke Breitfeller, Carolyn Rose
    [pdf, code/data]
  4. SIGDIAL 2019
  5. Exploring Numeracy in Word Embeddings
    Aakanksha Naik*, Abhilasha Ravichander*, Carolyn Rose, Eduard Hovy
    [pdf]
  6. ACL 2019
  7. Using Functional Schemas to Understand Social Media Narratives
    Xinru Yan, Aakanksha Naik, Yohan Jo, Carolyn Rose
    [pdf, code/data]
  8. Storytelling Workshop, ACL 2019
  9. Stress Test Evaluation for Natural Language Inference
    Aakanksha Naik*, Abhilasha Ravichander*, Norman Sadeh, Carolyn Rose, Graham Neubig
    Area Chair Favorite Paper [pdf, code/data]
  10. COLING 2018
  11. Extracting Personal Medical Events for User Timeline Construction using Minimal Supervision [pdf]
    Aakanksha Naik, Chris Bogart, Carolyn Rose
  12. BioNLP, ACL 2017
  13. Tackling Biomedical Text Summarization: OAQA at BioASQ 5B [pdf]
    Khyathi Chandu, Aakanksha Naik, Aditya Chandrasekar, Zi Yang, Niloy Gupta, Eric Nyberg
  14. BioNLP, ACL 2017
  15. A Computational Analyses of Online Support Group Discussions: Identifying Coping Strategies in LVAD Patients
    Aakanksha Naik*, Lisa Lohmueller*, Colleen McIlvennan, Dawn Christensen, Carolyn Rose, James Antaki
  16. Under revision for MDM
  17. Factors Affecting Healthcare Engagement by Patients with Severe Heart Failure: An Exploratory Analysis
    Lisa Lohmueller, Aakanksha Naik, Luke Breitfeller, Colleen McIlvennan, Manreet Kanwar, Srinivas Murali, Carolyn Rose, James Antaki
  18. Under review at MDM Policy and Practice

Projects

Thread Disentanglement for Multi-Party Conversations
Advisor: Prof Carolyn Rose
  • Developing a model for thread disentanglement in multi-party conversations leveraging utterance-level signals such as negotiation strategies displayed, and conversation-level signals such as inference links between context and utterance.
  • Focusing on disentanglement for transcripts of patient-doctor conversations as a first step towards building a summarization system for multi-party conversations.
August 2019 - now
Open-Domain Event Extraction
Advisor: Prof Carolyn Rose
  • Working on developing an event extractor which can be used across multiple domains with few training examples. We plan to leverage available data to generate more examples via distant supervision.
  • Developing this extractor with the aim of using it for event extraction from patient-doctor conversations.
August 2019 - now

**More details about past projects available in publications


Personal

In my free time, I enjoy reading, usually mystery and humour (especially British humour). I also love reading literary classics, short stories and fantasy. Though my list of favourite books keeps fluctuating, if you ask me for recommendations, I will probably suggest some of the following:

  • To Kill a Mockingbird: Harper Lee
  • A Thousand Splendid Suns: Khaled Hosseini
  • The Book Thief: Markus Zusak
  • Gone With the Wind: Margaret Mitchell
  • Little Women: Louisa May Alcott
  • Three Men in A Boat: Jerome K Jerome
  • The Importance of Being Earnest: Oscar Wilde
  • A Damsel in Distress, Big Money: PG Wodehouse
  • The Ransom of Red Chief, Mammon and the Archer, The Last Leaf, Brickdust Row, The Trimmed Lamp: O'Henry
  • The Lottery Ticket, Ward Number 6, The Lady and the Dog: Anton Chekhov
  • Murder on the Orient Express, The Murder of Roger Ackroyd, The Thirteen Problems: Agatha Christie

I often dabble in painting, sketching and calligraphy, which I first picked up during a summer workshop. You can check out some of my art here! I also like cooking and travelling and find both activities relaxing. Though I am not a huge TV enthusiast, I love watching Game of Thrones, The Good Place and Sherlock. I also enjoy listening to podcasts, especially Lore by Aaron Mahnke, which you should check out if you like creepy thrillers!