Skip to main content
Back to KTH start page

Ulme Wennberg

Profile picture of Ulme Wennberg

DOCTORAL STUDENT

Details

Address
LINDSTEDTSVÄGEN 24

Researcher


About me

Introduction

I am Ulme Wennberg, currently pursuing my doctoral research at KTH Royal Institute of Technology in Stockholm. My specialization lies in Machine Learning and Artificial Intelligence, with a keen focus on Natural Language Processing (NLP). Since 2020, I have been part of the Wallenberg AI, Autonomous Systems, and Software Program (WASP), working under the guidance of my supervisors Gustav Eje Henter and Jonas Beskow.

Academic and Research Pursuits

My role as a PhD student involves extensive research in machine learning with a concentration on advancing the field of NLP. My work has led to notable contributions in international conferences, including ACL 2021 in Bangkok and ICASSP 2022 in Singapore. One of my significant works, "Entity, Relation, and Event Extraction with Contextualized Span Representations", was published at EMNLP 2019 in Hong Kong.

To delve deeper into my research endeavors, you can visit my Google Scholar profile.

Professional Background

Before embarking on my PhD journey, I acquired industry experience at:

  • Microsoft: As a Software Engineer, where I developed advanced telephony software and chatbot solutions.
  • University of Washington: My research internship here involved setting new standards in NLP-based information extraction.
  • Amazon Alexa Prize: Contributed to NLU capabilities for an open-domain chatbot, achieving top-five recognition globally.

Academic Credentials

PhD in Artificial Intelligence (2020 - 2025) at KTH Royal Institute of Technology.

Graduate Exchange in Computer Science at the University of Washington.

MS in Machine Learning at KTH Royal Institute of Technology.

BS in Engineering Physics at KTH Royal Institute of Technology.

Teaching and Mentorship

As a teaching assistant in the Foundations of Machine Learning course at KTH, I share my knowledge and experience with budding AI professionals. My commitment to education extends to supervising master's theses, guiding the next generation of researchers.

Personal Interests

In addition to my academic pursuits, I co-founded Enkla Studier, a private tutoring initiative. My leadership role here and my direct involvement in tutoring reflect my dedication to educational growth and knowledge sharing.

Connect with Me

For collaborations, discussions, or inquiries about my research, feel free to contact me at .

MSc Thesis Supervision:

Celine Helgesson Hallström (Spring 2023)
Thesis Topic: Evaluating GPT-4's capabilities for self-evaluation
Additional Note: Supervising in collaboration with Schibsted Futures Lab

Johan Luhr (Spring 2023)
Thesis Topic: Modelling credit default swap spreads for illiquid instruments using transformer models

Gustavo Teodoro Döhler Beck (Spring 2021)
Thesis Topic: Created a framework for controllable speech synthesis
Winner of the SAIS Best AI Master’s Thesis, awarded annually by the Swedish AI Society
Additional Note: Turned thesis into a research article and got accepted into ICASSP 2022 in Singapore

Henrik Holm (Spring 2021)
Thesis Topic: Finetuned and evaluated various BERT-models on Ericssons internal data

Gustav Kjellberg (Spring 2021)
Thesis Topic: Explored federated learning training at SEB


Courses

Foundations of Machine Learning (DD1420), assistant | Course web