I am a Ph.D. candidate at the Institute for Logic, Language and Computation (ILLC) at the University of Amsterdam and permanent visiting of the School of Informatics at the University of Edinburgh. I am part of the EdinburghNLP group.
Under the supervision of Prof. Ivan Titov, my work focuses on Machine Reading Comprehension also know as Question Answering. More generally, I am interested in (semi-)supervised and unsupervised deep neural network applications in combination with reasoning and reinforcement methods to approach Natural Language Understanding.
- Apr 2021 Excited to announce Editing Factual Knowledge in Language Models an efficient tool to edit factual knowledge of language models!
- Mar 2021 Excited to announce Multilingual Autoregressive Entity Linking that extends Autoregressive Entity Retrieval in 100+ languages! Code and models here!
- Feb 2021 Invited talk at Microsoft Research (Cambridge, UK): Question-Answering: graph-based approaches and entity retrieval.
- Fev 2021 Invited talk at Twitter Research (Cambridge, UK): Graph Neural Networks meet NLP: exploiting structure and unstructured data.
- Jan 2021 Two paper accepted at ICLR2021 both as Spotlight (top 5%)! Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking and Autoregressive Entity Retrieval! Nice start of the year! 😎
- Oct 2020 Extremmely efficient retrieval using a language model and constrained beam search! A new pre-print: Autoregressive Entity Retrieval. On Fire! 🔥
- Oct 2020 Another intepretability work is ready! Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
- Sept 2020 How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking acceppted at EMNLP2020! A step forward in the difficult direction of understanding deep neural models.
- Sept 2020 Introducing KILT, a new unified benchmark for knowledge-intensive NLP tasks: 11 datasets, 5 task families, a unified knowledge source and interface. Very exicited to have contributied to such a big project.
- June 2020 A new fully reparameterizable location-scale distribution on hyper-spheres that doesn’t require rejection sampling! A new pre-print: The Power Spherical distribution
- May 2020 I will join Facebook AI Reseach (FAIR) in London for the Summer 2020. I will be working with Sebastian Riedel, Tim Rocktäschel, and Fabio Petroni
- May 2020 Finally my new pre-print is ready: How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking