I am a PhD student in the NLP research group at University of Helsinki, supervised by Mathias Creutz and Jörg Tiedemann. My thesis focuses on the vector-based representations of paraphrastic sentences, i.e., sentences that carry a similar meaning but are realised in different surface forms. More specifically, I focus on the tasks of paraphrase detection in noisy training data conditions and paraphrase generation for promoting variation in generated sequences.
I am especially interested in questions related to computational semantics, distributional semantics and applications such as Machine Translation.
Teemu Vahtola, Mathias Creutz, Jörg Tiedemann. It Is Not Easy To Detect Paraphrases: Analysing Semantic Similarity With Antonyms and Negation Using the New SemAntoNeg Benchmark. In Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. 2022.
Teemu Vahtola, Eetu Sjöblom, Jörg Tiedemann, and Mathias Creutz. Modeling Noise in Paraphrase Detection. In Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022).
Teemu Vahtola, Mathias Creutz, Eetu Sjöblom, and Sami Itkonen. Coping with Noisy Training Data Labels in Paraphrase Detection. In Proceedings of the 2021 EMNLP Workshop W-NUT : The Seventh Workshop on Noisy User-generated Text.
Eetu Sjöblom, Mathias Creutz, Teemu Vahtola. Grammatical Error Generation Based on Translated Fragments. In Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa) 2021.
You can contact me by email teemu.vahtola@helsinki.fi