no code implementations • 4 Jan 2022 • Ruben Cartuyvels, Graham Spinks, Marie-Francine Moens
Motivated by these insights, in this paper we argue that combining discrete and continuous representations and their processing will be essential to build systems that exhibit a general form of intelligence.
1 code implementation • COLING 2020 • Ruben Cartuyvels, Graham Spinks, Marie-Francine Moens
This paper proposes an iterative inference algorithm for multi-hop explanation regeneration, that retrieves relevant factual evidence in the form of text snippets, given a natural language question and its answer.
no code implementations • 7 Jul 2020 • Graham Spinks, Marie-Francine Moens
The paper proposes a novel technique for representing templates and instances of concept classes.
1 code implementation • NeurIPS 2020 • Dario Pavllo, Graham Spinks, Thomas Hofmann, Marie-Francine Moens, Aurelien Lucchi
A key contribution of our work is the encoding of the mesh and texture as 2D representations, which are semantically aligned and can be easily modeled by a 2D convolutional GAN.
no code implementations • 25 Sep 2019 • Graham Spinks, Marie-Francine Moens
We introduce a new deep learning technique that builds individual and class representations based on distance estimates to randomly generated contextual dimensions for different modalities.
no code implementations • 12 Jul 2019 • Graham Spinks, Marie-Francine Moens
This textual representation is decoded into a diagnosis and the associated textual justification that will help a clinician evaluate the outcome.
no code implementations • WS 2018 • Graham Spinks, Marie-Francine Moens
The method is illustrated on a medical dataset where the correct representation of spatial information and shorthands are of particular importance.
no code implementations • NAACL 2018 • Graham Spinks, Marie-Francine Moens
During training the input to the system is a dataset of captions for medical X-Rays.