Mete Mahammad Ismayilzada

About Me

I am currently a Ph.D. student at EPFL advised by prof.Antoine Bosselut at the NLP lab and a Research Assistant at Idiap Research Institute supervised by prof.Lonneke van der Plas at the Computation, Cognition and Language group.

Previously, I completed my MSc. in Data Science at EPFL where I did research on commonsense and causal reasoning. In the past, I have been a software/ML engineer, co-founder/entrepreneur and have 5+ years of experience in areas ranging from semiconductors to AI. I am also a proud University of Waterloo alumnus where I finished my Bachelor’s in Computer Science with distinction.

Research Interests

My interests broadly lie within the nature of robust artificial intelligence, language, creativity and mind. Currently, my research focuses on understanding, evaluating and improving creativity of AI systems. Additionally, I am investigating core abilities essential for creativity in AI, including reasoning, generalization, compositionality, abstraction, and analogy-making.

Publications

  1. Evaluating Creative Short Story Generation in Humans and Large Language Models
    Mete Ismayilzada, Claire Stevenson, Lonneke van der Plas
    Preprint
    [paper] [code]

  2. Creativity in AI: Progresses and Challenges
    Mete Ismayilzada, Debjit Paul, Antoine Bosselut, Lonneke van der Plas
    Preprint
    [paper] [code]

  3. Evaluating Morphological Compositional Generalization in Large Language Models
    Mete Ismayilzada, Defne Circi*, Jonne Sälevä*, Hale Sirin, Abdullatif Köksal, Bhuwan Dhingra, Antoine Bosselut, Lonneke van der Plas, Duygu Ataman
    Preprint
    [paper]

  4. DiffuCOMET: Contextual Commonsense Knowledge Diffusion
    Silin Gao, Mete Ismayilzada, Mengjie Zhao, Hiromi Wakaki, Yuki Mitsufuji, Antoine Bosselut
    ACL 2024
    [paper]

  5. Exploring Defeasibility in Causal Reasoning
    Shaobo Cui, Lazar Milikic, Yiyang Feng, Mete Ismayilzada, Debjit Paul, Antoine Bosselut, Boi Faltings
    ACL 2024 Findings
    [paper]

  6. REFINER: Reasoning Feedback on Intermediate Representations
    Debjit Paul, Mete Ismayilzada, Maxime Peyrard, Beatriz Borges, Antoine Bosselut, Robert West, Boi Faltings
    EACL 2024
    [paper]

  7. Could ChatGPT be an Engineer? Evaluating Higher Education Vulnerability to AI Assistants
    Beatriz Borges, Negar Foroutan, Deniz Bayazit, Syrielle Montariol, Mohammadreza Banaei, […], Mete Ismayilzada, Debjit Paul, Tanja Käser, Antoine Bosselut
    AI for Education: Bridging Innovation and Responsibility, AAAI 2024
    [paper]

  8. CRoW: Benchmarking Commonsense Reasoning in Real-World Tasks
    Mete Ismayilzada, Debjit Paul*, Syrielle Montariol*, Mor Geva, Antoine Bosselut
    EMNLP 2023 (Oral)
    [paper] [website]

  9. kogito: A Commonsense Knowledge Inference Toolkit.
    Mete Ismayilzada, Antoine Bosselut.
    EACL System Demonstrations Track, 2023 (Poster)
    [paper] [code] [docs]

Education

  1. Ph.D in Computer Science. EPFL. Lausanne, Switzerland. 2023-Present.
  2. Master’s in Data Science. EPFL. Lausanne, Switzerland. 2021-2023.
  3. Bachelor’s in Computer Science. University of Waterloo. Waterloo, Canada. 2012-2016.
  4. Bachelor’s in Computer Science. Baku State University. Baku, Azerbaijan. 2010-2012 (incomplete)

Research Experience

  1. Research Assistant. IDIAP. Martigny, Switzerland. 2023-Present.
  2. Research Assistant. Artificial Intelligence Lab. EPFL. Fall 2022.
  3. Research Intern. Sony Inc. Tokyo, Japan. Summer 2022.

Teaching Experience

  1. Teaching Assistant. Modern NLP. EPFL. Spring 2024.
  2. Student Teaching Assistant. Modern NLP. EPFL. Spring 2023.
  3. Student Teaching Assistant. Applied Data Analysis. EPFL. Fall 2022.

Engineering Experience

  1. Software/ML Engineer. integrate.ai. Toronto, Canada. 2019-2021.
  2. Co-founder/CTO. Destin AI. Toronto, Canada. 2017-2019.
  3. Software Engineer. Bluecat Networks. Toronto, Canada. 2017-2018.
  4. Software Engineer. Cisco Systems. Ottawa, Canada. 2016-2017.
  5. Software Engineer Intern. Freescale Semiconductors. Ottawa, Canada. Summer 2015.
  6. Software Developer Intern. TD Securities. Toronto, Canada. Summer 2014.
  7. Software Developer Intern. Azercell MMC. Baku, Azerbaijan. Summer 2013.

Service

  1. Reviewer. SEM. Apr 2024.
  2. Reviewer. ACL ARR. Feb 2024.
  3. Volunteer. Perspectives on AI Symposiums. IDIAP. Feb 2024.
  4. Reviewer. EACL Student Research Workshop. Jan 2024.

Projects

  1. Parrot. Educational toolkit for interaction with OpenAI models. [code]
  2. kogito. A Python Commonsense Knowledge Inference Toolkit. [docs] [code]
  3. mini-pytorch. From scratch implementation of (mini) pytorch autograd and several core modules. [code]
  4. Words and Personalities. Personality analysis based on quotes. [demo] [code]
  5. Differentially Private Random Forest Classifier. [code]

Blogposts

  1. From Words to Vectors. Towards Data Science. Medium. [link]

Hobbies

In my free time, I like to read, road bike, play piano and dombra and watch (usually long) international art films.

Contact

  1. mismayilza [at] gmail [dot] com
  2. github
  3. linkedin

Fin