Ákos Kádár

Hey there,

You’ve reached the professional page of Ákos Kádár. I’m currently located in Luxembourg. Here is my short bio:

I’m a researcher and developer in Machine Learning with the majority of my experience in Natural Language Processing. Currently working on music generation with the R&D team of AIVA bringing powerful generative models of music into production.

Before moving onto music I worked at explosion on the popular NLP library spaCy and thinc the machine learning library powering it. I’m grateful that I was one of the lucky ones who had the opportunity to make a living writing open-source code.

Before coming back to Europe I’ve spent some lovely time in Canada and I’m still looking to go back if life turns out that way. I was a Machine Learning Researcher at Borealis AI in Toronto working mainly on semantic parsing from English to SQL. During this time we’ve published a nice three part tutorial on syntactic parsing with Simon J.D. Prince if you are interested. Prior to Borealis I was a researcher at Samsung AI Toronto, where we published on semantic parsing with graph grammars and dependency parsing with tree embeddings. I also did an internship in Microsoft Research Montreal, where we developed the procedural text-based adventure game generator TextWorld mainly with Marc-Alexandre Côté.

Before starting in the industry I did my PhD at the department of Cognitive Science & Artificial Intelligence at Tilburg University under the supervision of Grzegorz Chrupala and Afra Alishahi. Wrote dissertation on learning multi-modal and multilingual sentence representations.

Publications

Peng Xu, Wenjie Zi, Hamidreza Shahidi, Ákos Kádár, Keyi Tang, Wei Yang, Jawad Ateeq, Harsh Barot, Meidan Alon, Yanshuai Cao
Turing: an Accurate and Interpretable Multi-Hypothesis Cross-Domain Natural Language Database Interface
ACL 2021 Demo track | Website

Ákos Kádár, Lan Xiao, Mete Kemertas, Federico Fancellu, Allan Jepson, Afsaneh Fazly
Dependency parsing with structure preserving embeddings
EACL 2021

Judit Ács, Ákos Kádár, Andras Kornai
Subword Pooling Makes a Difference
EACL 2021 | Code

Chris Emmery, Ákos Kádár, Grzegorz Chrupała
Adversarial Stylometry in the Wild: Transferable Lexical Substitution Attacks on Author Profiling
EACL 2021 | Code

Federico Fancellu, Ákos Kádár, Ran Zhang, Afsaneh Fazly
Accurate Polyglot Semantic Parsing With DAG Grammars
EMNLP Findings 2020

Chris Emmery, Ákos Kádár, Travis J. Wiltshire, Andrew T. Hendrickson
Towards Replication in Computational Cognitive Modeling: A Machine Learning Perspective
Computational Brain & Behavior 2019

Enrique Manjavacas, Ákos Kádár, and Mike Kestemont
Improving Lemmatization of Non-Standard Languages with Joint Learning
NAACL 2019

Grzegorz Chrupała, Lieke Gelderloos, Ákos Kádár and Afra Alishahi
On the difficulty of a distributional semantics of spoken language
Proceedings of the Society for Computation in Linguistics 2019

Ákos Kádár, Desmond Elliott, Marc-Alexandre Côté, Grzegorz Chrupała and Afra Alishahi
Lessons learned in multilingual grounded language learning
EMNLP/ConLL 2018 | Code

Marc-Alexandre Côté, Ákos Kádár, Xingdi Yuan, Ben Kybartas, Tavian Barnes, Emery Fine, James Moore, Matthew Hausknecht, Layla El Asri, Mahmoud Adada, Wendy Tay, Adam Trischler
TextWorld: A Learning Environment for Text-based Games
Computer Games Workshop at ICML/IJCAI 2018 | Code

Ákos Kádár, Marc-Alexandre Côté, Grzegorz Chrupała and Afra Alishahi
Revisiting the Hierarchical Multiscale LSTM
COLING 2018

Emiel van Miltenburg, Ákos Kádár, Ruud Koolen and Emiel Krahmer
DIDEC: The Dutch Image Description and Eye-tracking Corpus
COLING 2018 | Website

Thiago Castro Ferreira, Diego Moussallem, Ákos Kádár, Sander Wubben and Emiel Krahmer
NeuralREG: An end-to-end approach to referring expression generation
ACL 2018 | Code

Samira Ebrahimi Kahou, Adam Atkinson, Vincent Michalski, Ákos Kádár, Adam Trischler, Yoshua Bengio
FigureQA: An Annotated Figure Dataset for Visual Reasoning
ICLR 2018 Workshop

Desmond Elliott, Ákos Kádár
Imagination improves Multimodal Translation
IJCNLP 2017 | Code

Ákos Kádár, Grzegorz Chrupała, Afra Alishahi
Representation of linguistic form and function in recurrent neural networks
Computational Linguistics 2017 | Code

Ákos Kádár, Grzegorz Chrupała, Afra Alishahi
Linguistic Analysis of multi-modal Recurrent Neural Networks
EMNLP VL’15 (poster presentation)

Grzegorz Chrupała, Ákos Kádár, Afra Alishahi
Learning language through pictures
ACL-IJCNLP 2015 | Code

Ákos Kádár, Grzegorz Chrupała, Afra Alishahi
Learning word meanings from images of natural scenes
TAL 2016 | Code

Master Thesis

Ákos Kádár
Grounded Learning for Source Code Component Retrieval
Code