Giorgos Vernikos

prof_pic.jpg

Hello! I am a final-year Ph.D. student at the Electrical Engineering department at EPFL and a research assistant at HEIG-VD, in Switzerland. My Ph.D. advisor is Andrei Popescu-Belis. I am interested in machine translation and large language models. I also have a growing interest in machine translation evaluation.

During my Ph.D. I interned at Google in the Google Assistant team in Zurich, Switzerland. I also did an internship at Amazon Web Services (AWS) in Santa Clara, CA, working with the Amazon Translate team.

Before coming to Switzerland, I obtained my diploma (combined BEng and MEng) in Electrical and Computer Engineering and my MSc in Data Science and Machine Learning from the National Technical University of Athens in Athens, Greece. In my Master’s thesis, I worked on Adversarial Fine-Tuning of Pretrained Language Models under the supervision of Andreas Stafylopatis. During that time I was also working as a Machine Learning Engineer at DeepSea Technologies.

news

May 16, 2024 Excited to announce that our paper Don’t Rank, Combine! Combining Machine Translation Hypotheses Using Quality Estimation has been accepted at the main conference of ACL 2024!
Feb 21, 2024 Invited talks at IST & Unbabel Seminars and Microsoft MT reading group!
Jan 18, 2024 My Google internship project Small Language Models Improve Giants by Rewriting Their Outputs has been accepted at EACL 2024!
Jan 16, 2024 The preprint for our work Don’t Rank, Combine! Combining Machine Translation Hypotheses Using Quality Estimation is available on Arxiv!
Jul 14, 2023 Invited talk at Archimedes 2023 Summer NLP Workshop in Athens!

selected publications

  1. ACL
    Don’t Rank, Combine! Combining Machine Translation Hypotheses Using Quality Estimation
    G. Vernikos, and A. Popescu-Belis
    In Proceedings of the 62st Annual Meeting of the Association for Computational Linguistics, 2024
  2. EACL

    🎤 Oral

    Small Language Models Improve Giants by Rewriting Their Outputs
    In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, 2024
  3. WMT

    🎤 Oral

    Embarrassingly Easy Document-Level MT Metrics: How to Convert Any Pretrained Metric Into a Document-Level Metric
    G. VernikosB. ThompsonP. Mathur, and M. Federico
    In Proceedings of the Seventh Conference on Machine Translation, 2022
  4. EMNLP
    Subword Mapping and Anchoring across Languages
    G. Vernikos, and A. Popescu-Belis
    In Findings of the Association for Computational Linguistics: EMNLP 2021, 2021
  5. EMNLP

    🎤 Oral

    Active Learning by Acquiring Contrastive Examples
    K. MargatinaG. VernikosL. Barrault, and N. Aletras
    In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, 2021