Ramakanth Pasunuru

I am a fifth year PhD student in Computer Science, UNC Chapel Hill. I work with Prof. Mohit Bansal. My research focuses on developing knowledgeable language generation models with multimodal context that incorporate generalizable semantic skills via multi-task learning, reinforcement learning, and continual learning methods.

Recently, I worked with Markus Dreyer as a research intern at Amazon. Previously, I worked with Asli Celikyilmaz and Michel Galley as a research intern at MSR. I also worked with David Rosenberg as a research intern at Bloomberg.

I graduated from Electrical Engineering with specialization in Instrumentation and Signal Processing, IIT Kharagpur in 2015, where I worked with Prof. Anirban Mukherjee and was part of the Signal Processing and Machine Learning Research Group.

I have also worked with Prof. Oliver Schulte at Simon Fraser University, where I was a Mitacs Globalink Ambassador Intern during the summer of 2014.

I am grateful to be supported by a Microsoft PhD Fellowship.


  • One paper accepted at NAACL 2021 -- New
  • One paper accepted at AAAI 2021 -- New
  • Two papers accepted at EMNLP 2020 -- New
  • One paper accepted at AAAI 2020
  • One paper accepted at ACL 2019
  • Doing summer internship at MSR with Asli Celikyilmaz and Michel Galley
  • Two papers accepted at NAACL 2019
  • Serving as publicity/sponsorship chair for CoNLL 2019
  • Awarded the Microsoft Research PhD Fellowship (Finalist for Facebook Research PhD Fellowship)
  • Our paper "Game-Based Video-Context Dialogue" accepted to EMNLP 2018 (pdf)
  • Doing summer internship at Bloomberg with David Rosenberg
  • Our paper "Dynamic Multi-Level Multi-Task Learning for Sentence Simplification" accepted to COLING 2018 (pdf). Also, selected as Area Chair Favorites
  • Our paper "Soft Layer-Specific Multi-Task Summarization with Entailment and Question Generation" accepted to ACL 2018 (pdf)
  • Our paper "Multi-Reward Reinforced Summarization with Saliency and Entailment" accepted to NAACL 2018 (pdf)
  • Our demo "Interactive-Length Multi-Task Video Captioning with Cooperative Feedback" was accepted to NIPS 2017 demo track
  • Our paper "Reinforced Video Captioning with Entailment Rewards" accepted to EMNLP 2017 (pdf)
  • Contributed Talk: Our paper "Towards Improving Abstractive Summarization via Entailment Generation" accepted to Workshop on Summarization Frontiers, EMNLP 2017 (pdf)
  • Thanks for ACL Don and Betty Walker Scholarship
  • Outstanding Paper Award for "Multi-Task Video Captioning with Video and Entailment Generation"
  • Our paper "Multi-Task Video Captioning with Video and Entailment Generation" accepted to ACL 2017 (pdf)