• Publications
  • Influence
Author Masking through Translation
This paper documents the approach adopted by our team for Author Masking Task in PAN 2016. Expand
Repurposing Entailment for Multi-Hop Question Answering Tasks
We introduce Multee, a general architecture that can effectively use entailment models for multi-hop QA tasks. Expand
Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning
We formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. Expand
DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering
We introduce DeFormer, a decomposed transformer, which substitutes the full self-attention with question-wide and passage-wide self-Attentions in the lower layers, which in turn enables pre-computing passage representations reducing runtime compute drastically. Expand
Controlling Information Aggregation for Complex Question Answering
This paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Expand
Measuring and Reducing Non-Multifact Reasoning in Multi-hop Question Answering
The measurement of true progress in multihop question-answering has been muddled by the strong ability of models to exploit artifacts and other reasoning shortcuts. Expand
Doubly commuting invariant subspaces for representations of product systems of $C^*$-correspondences.
We obtain a Shimorin-Wold-type decomposition for a doubly commuting covariant representation of a product system of $C^*$-correspondences. This extends a recent Wold-type decomposition by Jeu andExpand
Noise Correction in Pairwise Document Preferences for Learning to Rank
This paper proposes a way of correcting noise in the training data for Learning to Rank. Expand
What Ingredients Make for an Effective Crowdsourcing Protocol for Difficult NLU Data Collection Tasks?
Crowdsourcing is widely used to create data for common natural language understanding tasks. Despite the importance of these datasets for measuring and refining model understanding of language, thereExpand