Paolo Gajo

Profile

PhD student in NLP at the University of Bologna, focusing on gastronomy research through LLMs. Currently a visiting researcher at the Hypermatrix group at Dalhousie University, Canada, working on graph neural networks for text-to-graph models. Personal research includes ASR and seq2seq models for automatic subtitling for the deaf and hard of hearing.

Publications

• Paolo Gajo and Alberto Barrón-Cedeño. "On Cross-Language Entity Label Projection and Recognition." In CLiC-it 2024: 10th Italian Conference on Computational Linguistics, 2024.
• Paolo Gajo, Luca Giordano, and Alberto Barron-Cedeño. "UniBO at CheckThat! 2024: Multi-lingual and multi-label persuasion technique detection in news with data augmentation and sequence-token classifiers." In CLEF 2024, Grenoble, France, 2024.
• Paolo Gajo, Arianna Muti, Katerina Korre, Silvia Bernardini, and Alberto Barrón-Cedeño. "On the Identification and Forecasting of Hate Speech in Inceldom." In 14th International Conference on Recent Advances in Natural Language Processing, Varna, Bulgaria, 2023.
• Paolo Gajo, Silvia Bernardini, Adriano Ferraresi, Barrón-Cedeño Alberto, et al. "Hate speech detection in an Italian incel forum using bilingual data for pre-training and fine-tuning." In CLiC-it 2023, Venice, Italy, 2023.
• Aikaterini Korre, Paolo Gajo, and Alberto Barrón-Cedeño. "Hate Speech According to the Law: An Analysis for Effective Detection." In 31st International Conference on Computational Linguistics (COLING 2025), 2025 (to appear).

Projects

• Automatic Subtitle Segmenter (2023): Fine-tuned a T5 model on a dataset created from subtitling work to build an automatic subtitle segmenter for audiovisual content. Code: GitHub [https://github.com/paolo-gajo/subsplitter]

Skills

• Programming: Python, R, Bash
• Libraries: PyTorch, NumPy, Transformers, Pandas, Sci-Kit Learn, SpaCy, Selenium, BeautifulSoup
• Software: Git, VS Code, LaTeX, Overleaf, Slurm, Sketch Engine, AntConc, RegEx, Excel, Label Studio, Trados, MemoQ, Subtitle Edit, Aegisub
• Text Processing: RegEx, Sketch Engine, AntConc, BootCaT
• Translation: Trados, MemoQ, MateCat, MultiTerm, Subtitle Edit, Aegisub

Education

• Dalhousie University, Halifax, Canada
  Visiting @ Hypermatrix (Sep 2024 – Sep 2025)
  Researching graph neural networks and text-to-graph models.
  Courses: MATH/STAT 2060 - Intro Probability & Statistics, CSCI 3151 - Foundations of Machine Learning, CSCI 4158 - NLP with Deep Learning, CSCI 6516 - Deep Learning.

• University of Bologna, Forlì, Italy
  PhD in Natural Language Processing (2023 – Present)
  3-year PhD on the use of NLP for studying gastronomy.

• University of Bologna, Forlì, Italy
  MA in Specialized Translation (110/110 Summa Cum Laude, 2021 – 2023)
  Focus on translation technologies and translation in English, Spanish, and Italian.
  Thesis: "This Is My Cope: Identification and Forecasting of Hate Speech in Inceldom." Code: GitHub [https://github.com/paolo-gajo/incel-thesis]

• Ca’ Foscari University, Treviso, Italy
  BA in Linguistic and Cultural Mediation (110/110 Summa Cum Laude, 2018 – 2021)
  Focus on translation and interpreting in English, Spanish, and Italian.
  Thesis: "Practical Application of Subtitling Guidelines - Intralingual and Interlingual Subtitling of the Short Film 'Group B'"