Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Rethinking Evaluation Practices in Visual Question Answering: A Case Study on Out-of-Distribution Generalization


May 24, 2022
Aishwarya Agrawal, Ivana Kajić, Emanuele Bugliarello, Elnaz Davoodi, Anita Gergely, Phil Blunsom, Aida Nematzadeh

* Aishwarya, Ivana, Emanuele and Aida had equal first author contributions. Elnaz and Anita had equal contributions. Aida and Aishwarya had equal senior contributions. Paper has 29 pages, 8 figures, 15 tables 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

StreamingQA: A Benchmark for Adaptation to New Knowledge over Time in Question Answering Models


May 23, 2022
Adam Liška, Tomáš Kočiský, Elena Gribovskaya, Tayfun Terzi, Eren Sezener, Devang Agrawal, Cyprien de Masson d'Autume, Tim Scholtes, Manzil Zaheer, Susannah Young, Ellen Gilsenan-McMahon, Sophia Austin, Phil Blunsom, Angeliki Lazaridou


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Revisiting the Compositional Generalization Abilities of Neural Sequence Models


Mar 14, 2022
Arkil Patel, Satwik Bhattamishra, Phil Blunsom, Navin Goyal

* ACL 2022 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale


Mar 01, 2022
Laurent Sartran, Samuel Barrett, Adhiguna Kuncoro, Miloš Stanojević, Phil Blunsom, Chris Dyer

* 24 pages, 9 figures, 3 tables and 1 algorithm 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Relational Memory Augmented Language Models


Jan 24, 2022
Qi Liu, Dani Yogatama, Phil Blunsom

* Accepted to TACL, pre MIT Press publication version 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

A Systematic Investigation of Commonsense Understanding in Large Language Models


Oct 31, 2021
Xiang Lorraine Li, Adhi Kuncoro, Cyprien de Masson d'Autume, Phil Blunsom, Aida Nematzadeh


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Pretraining the Noisy Channel Model for Task-Oriented Dialogue


Mar 18, 2021
Qi Liu, Lei Yu, Laura Rimell, Phil Blunsom

* Accepted to TACL, pre MIT Press publication version 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Pitfalls of Static Language Modelling


Feb 03, 2021
Angeliki Lazaridou, Adhiguna Kuncoro, Elena Gribovskaya, Devang Agrawal, Adam Liska, Tayfun Terzi, Mai Gimenez, Cyprien de Masson d'Autume, Sebastian Ruder, Dani Yogatama, Kris Cao, Tomas Kocisky, Susannah Young, Phil Blunsom


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Mutual Information Constraints for Monte-Carlo Objectives


Dec 01, 2020
Gábor Melis, András György, Phil Blunsom

* 32 pages, 29 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

The Struggles of Feature-Based Explanations: Shapley Values vs. Minimal Sufficient Subsets


Sep 23, 2020
Oana-Maria Camburu, Eleonora Giunchiglia, Jakob Foerster, Thomas Lukasiewicz, Phil Blunsom


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email
1
2
3
4
5
6
7
>>