Search Results for author: Jeff Mitchell

Found 14 papers, 2 papers with code

Generalisation in Neural Networks Does not Require Feature Overlap

no code implementations4 Jul 2021 Jeff Mitchell, Jeffrey S. Bowers

That shared features between train and test data are required for generalisation in artificial neural networks has been a common assumption of both proponents and critics of these models.

Priorless Recurrent Networks Learn Curiously

no code implementations COLING 2020 Jeff Mitchell, Jeffrey Bowers

Recently, domain-general recurrent neural networks, without explicit linguistic inductive biases, have been shown to successfully reproduce a range of human language behaviours, such as accurately predicting number agreement between nouns and verbs.

Language Acquisition

Symmetry and Systematicity

no code implementations25 Sep 2019 Jeff Mitchell, Jeff Bowers

We argue that symmetry is an important consideration in addressing the problem of systematicity and investigate two forms of symmetry relevant to symbolic processes.

UCL Machine Reading Group: Four Factor Framework For Fact Finding (HexaF)

no code implementations WS 2018 Takuma Yoneda, Jeff Mitchell, Johannes Welbl, Pontus Stenetorp, Sebastian Riedel

In this paper we describe our 2nd place FEVER shared-task system that achieved a FEVER score of 62. 52{\%} on the provisional test set (without additional human evaluation), and 65. 41{\%} on the development set.

Information Retrieval Natural Language Inference +1

Jack the Reader -- A Machine Reading Framework

1 code implementation ACL 2018 Dirk Weissenborn, Pasquale Minervini, Isabelle Augenstein, Johannes Welbl, Tim Rockt{\"a}schel, Matko Bo{\v{s}}njak, Jeff Mitchell, Thomas Demeester, Tim Dettmers, Pontus Stenetorp, Sebastian Riedel

For example, in Question Answering, the supporting text can be newswire or Wikipedia articles; in Natural Language Inference, premises can be seen as the supporting text and hypotheses as questions.

Information Retrieval Language understanding +5

Jack the Reader - A Machine Reading Framework

2 code implementations20 Jun 2018 Dirk Weissenborn, Pasquale Minervini, Tim Dettmers, Isabelle Augenstein, Johannes Welbl, Tim Rocktäschel, Matko Bošnjak, Jeff Mitchell, Thomas Demeester, Pontus Stenetorp, Sebastian Riedel

For example, in Question Answering, the supporting text can be newswire or Wikipedia articles; in Natural Language Inference, premises can be seen as the supporting text and hypotheses as questions.

Language understanding Link Prediction +4

Extrapolation in NLP

no code implementations WS 2018 Jeff Mitchell, Pasquale Minervini, Pontus Stenetorp, Sebastian Riedel

We argue that extrapolation to examples outside the training space will often be easier for models that capture global structures, rather than just maximise their local fit to the training data.

Behavior Analysis of NLI Models: Uncovering the Influence of Three Factors on Robustness

no code implementations NAACL 2018 Vicente Ivan Sanchez Carmona, Jeff Mitchell, Sebastian Riedel

Natural Language Inference is a challenging task that has received substantial attention, and state-of-the-art models now achieve impressive test set performance in the form of accuracy scores.

Natural Language Inference

Reduce, Reuse, Recycle: New uses for old QA resources

no code implementations22 Apr 2018 Jeff Mitchell, Sebastian Riedel

We investigate applying repurposed generic QA data and models to a recently proposed relation extraction task.

Relation Extraction Slot Filling

Cannot find the paper you are looking for? You can Submit a new open access paper.