| layout | title | permalink |
|---|---|---|
home |
Martin Funkquist - AI Researcher |
/ |
{: width="300" }
I am Martin Funkquist, currently a PhD student at Linköping University. My research is within the field of AI, with a focus on sequential decision making and learning. I am interested in creating autonomous agents that can autonomously learn how to act in a complext environment. Our current approaches involve combining Deep Learning and Symbolic AI that can learn symbolic models of the world, which can be used for planning and decision making.
I did both my Bachelor's degree in Simulation Technology and Virtual Design and Master's degree in Computer Science at KTH. After working a few years as a Software Engineering consultant and doing a research internship at TU Darmstadt I am now a PhD student in the Machine Reasoning Lab and work on the Representation Learning for Acting and Planning project.
Martin Funkquist, Simon Ståhlberg, Hector Geffner (2024) Learning to Ground Existentially Quantified Goals In Proceedings of the Twenty-First International Conference on Principles of Knowledge Representation and Reasoning (KR 2024)
Martin Funkquist, Ilia Kuznetsov, Yufang Hou, Iryna Gurevych (2023) CiteBench: A benchmark for Scientific Citation Text Generation In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023.
[PDF]({{ site.baseurl }}/assets/papers/funkquist-et-al-2023.pdf) [Citation]({{ site.baseurl }}/assets/citations/funkquist-et-al-2023.bib)
Martin Funkquist (2021) Combining sentence and table evidence to predict veracity of factual claims using TaPaS and RoBERTa In Proceedings of the Fourth Workshop on Fact Extraction and VERification (FEVER), 2021.
[PDF]({{ site.baseurl }}/assets/papers/funkquist-2021.pdf) [Citation]({{ site.baseurl }}/assets/citations/funkquist-2021.bib)
The easiest way to contact me is via email: firstname.lastname@liu.se
You can also find me on different social media platforms:
{% include social-media-links.html %}
This work is open sourced under the Apache License, Version 2.0.
Copyright 2019 Tao He.