Hi! I am currently a Masters student in the department of Computer Science, in Reasoning and Learning Lab at McGill University, supervised by Jackie Chi Kit Cheung.

I am generally interested in Natural Language Generation, though my current research is focused on single document summarisation and their evaluation measures. Evaluation metrics such as rouge are the current norms and have limitations, most of my energy is currently going towards changing this norm.

The remaining energy goes into thought experiments like, can summarisation be solved without solving problems in Natural Language Understanding? Can we leverage knowledge from NLU tasks? or is supervised approach ideal for summarization? Do humans learn to summarise or they learn to understand?