TopiOCQA: Open-domain Conversational Question Answering with Topic Switching

Autor: Vaibhav Adlakha, Shehzaad Dhuliawala, Kaheer Suleman, Harm de Vries, Siva Reddy

Publikováno v: Transactions of the Association for Computational Linguistics, Vol 10, Pp 468-483 (2022)

AbstractIn a conversational question answering scenario, a questioner seeks to extract information about a topic through a series of interdependent questions and answers. As the conversation progresses, they may switch to related topics, a phenomenon

Externí odkaz: https://doaj.org/article/b6d9583b08d54f0db0c68948315bdbd5

Uni- and multivariate probability density models for numeric subgroup discovery

Autor: Arno Knobbe, Marvin Meeng, Siegfried Nijssen, Peter A. Flach, Harm de Vries

Publikováno v: Intelligent Data Analysis. 24:1403-1439

Subgroup Discovery is a supervised, exploratory data mining paradigm that aims to identify subsets of a dataset that show interesting behaviour with respect to some designated target attribute. The way in which such distributional differences are qua

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::4598b4a628e1f4ba77b8e0f2654c61bd
https://doi.org/10.3233/ida-194719

DuoRAT: Towards Simpler Text-to-SQL Models

Autor: Raymond Li, Chris Pal, Torsten Scholak, Dzmitry Bahdanau, Harm de Vries

Publikováno v: NAACL-HLT

Recent neural text-to-SQL models can effectively translate natural language questions to corresponding SQL queries on unseen databases. Working mostly on the Spider dataset, researchers have proposed increasingly sophisticated solutions to the proble

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::01c78105f7b04304a72305b37833d1bc
http://arxiv.org/abs/2010.11119

Visual Reasoning with Multi-hop Feature Modulation

Autor: Florian Strub, Harm de Vries, Philippe Preux, Olivier Pietquin, Aaron Courville, Ethan Perez, Mathieu Seurin, Jérémie Mary

Publikováno v: ECCV 2018-15th European Conference on Computer Vision
ECCV 2018-15th European Conference on Computer Vision, Sep 2018, Munich, Germany. pp.808-831
Computer Vision – ECCV 2018 ISBN: 9783030012274
ECCV (5)

Recent breakthroughs in computer vision and natural language processing have spurred interest in challenging multi-modal tasks such as visual question-answering and visual dialogue. For such tasks, one successful approach is to condition image-based

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::34b3312b9ed95d2f3197c78c2827505a
https://hal.archives-ouvertes.fr/hal-01927811

FiLM: Visual Reasoning with a General Conditioning Layer

Autor: Ethan Perez, Florian Strub, Harm De Vries, Vincent Dumoulin, Aaron Courville

Publikováno v: AAAI Conference on Artificial Intelligence
AAAI Conference on Artificial Intelligence, Feb 2018, New Orleans, United States

We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. W

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5a401ce52eb27fd8dc704ef00d1df852
https://hal.inria.fr/hal-01648685