Search Results for author: Mikael Kågebäck

Found 5 papers, 3 papers with code

Disentangled activations in deep networks

no code implementations ICLR 2018 Mikael Kågebäck, Olof Mogren

Deep neural networks have been tremendously successful in a number of tasks.

Disentangled Representations for Manipulation of Sentiment in Text

no code implementations22 Dec 2017 Maria Larsson, Amanda Nilsson, Mikael Kågebäck

The ability to change arbitrary aspects of a text while leaving the core message intact could have a strong impact in fields like marketing and politics by enabling e. g. automatic optimization of message impact and personalized language adapted to the receiver's profile.

Query-Based Abstractive Summarization Using Neural Networks

2 code implementations17 Dec 2017 Johan Hasselqvist, Niklas Helmertz, Mikael Kågebäck

In this paper, we present a model for generating summaries of text documents with respect to a query.

Abstractive Text Summarization

Learning to Play Guess Who? and Inventing a Grounded Language as a Consequence

1 code implementation10 Nov 2016 Emilio Jorge, Mikael Kågebäck, Fredrik D. Johansson, Emil Gustavsson

The images from the game provide a non trivial environment for the agents to discuss and a natural grounding for the concepts they decide to encode in their communication.

Image Retrieval

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