Preksha Nema

The heads hypothesis: A unifying statistical approach towards understanding multi-headed attention in BERT
On the Importance of Local Information in Transformer Based Models
On the weak link between importance and prunability of attention heads
Towards Interpreting BERT for Reading Comprehension Based QA
Towards Transparent and Explainable Attention Models
Let's Ask Again: Refine Network for Automatic Question Generation
Let’s Ask Again: Refine Network for Automatic Question Generation
ElimiNet: A Model for Eliminating Options for Reading Comprehension with Multiple Choice Questions
Generating Descriptions from Structured Data Using a Bifocal Attention Mechanism and Gated Orthogonalization
Towards a Better Metric for Evaluating Question Generation Systems
Diversity driven attention model for query-based abstractive summarization