Elamur Research⌕
☰
home / papers / 2026 / GRUPPETTOZZO at MultiPRIDE: Detecting LGBTQ+ Reclamatory Intent via Context-Aware Transformers

GRUPPETTOZZO at MultiPRIDE: Detecting LGBTQ+ Reclamatory Intent via Context-Aware Transformers

Paper · 2026 · Federico Traina, Alessandro Santoro, Gabriele Greco et al.

Summary

The GRUPPETTOZZO team (University of Palermo) built a system for the MultiPRIDE shared task at EVALITA 2026. The task asks whether an LGBTQ+-related slur in a tweet is used to reclaim it rather than to denigrate. Task A uses only the tweet text, in Italian, Spanish and English. Task B adds the author's user bio, in Italian and Spanish. Both tasks also have multilingual variants.

The team fine-tuned encoder-only transformers in two setups: one monolingual model per language, and a single XLM-RoBERTa model. The tweet and bio are joined into one input separated by [SEP]. A custom head combines CLS, mean and max pooling and passes the result to a small classifier.

To handle the small, imbalanced data, they used back-translation augmentation and compared weighted and focal losses. They also compared three text-cleaning levels.

The system beat the baseline in 6 of 10 subtasks and ranked first in subtask B2 (Spanish with bio), with macro F1 0.7304. Error analysis shows that recognising reclamation depends on pragmatics, not just vocabulary. Irony, discussion about slurs, and pride hashtags remain difficult.

Key points

  • Data: about 1,000 labelled tweets per language, split 80/20 with stratification. Back-translation used MarianMT/OPUS-MT, with English as the pivot for Italian and Spanish and Spanish as the pivot for English. Two strategies were tested: augmenting all samples or only the positive (minority) class.
  • Backbones:
    • Italian: umBERTo fine-tuned for hate/offensiveness (lupobricco).
    • English: BERT base uncased.
    • Spanish: BETO.
    • Multilingual runs: XLM-RoBERTa.
  • Pooling: the hybrid strategy concatenates CLS, mean and max vectors (2304 dimensions). It gave the best overall results.
  • Training choices: focal loss improved robustness from development to test, except in A3. Keeping emojis as text (demojization) was preferred over removing emojis and stopwords.
  • Effect of the bio: with monolingual backbones, adding the bio improved scores over task A. XLM-RoBERTa exceeded 0.90 macro F1 on task B in development, but on test it made poor use of the bio.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 works

Cite

Cite as Federico Traina, Alessandro Santoro, Gabriele Greco, Irene Siragusa and Roberto Pirrone. 2026. GRUPPETTOZZO at MultiPRIDE: Detecting LGBTQ+ Reclamatory Intent via Context-Aware Transformers. Proceedings of the Ninth Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2026).

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{traina2026gruppettozzo,
  title = {{GRUPPETTOZZO at MultiPRIDE: Detecting LGBTQ+ Reclamatory Intent via Context-Aware Transformers}},
  author = {Federico Traina, Alessandro Santoro, Gabriele Greco, Irene Siragusa and Roberto Pirrone},
  booktitle = {Proceedings of the Ninth Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2026)},
  year = {2026},
  url = {https://aclanthology.org/2026.evalita-1.24/}
}

References

18 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. C. Ferrando, L. Draetta, M. Madeddu, M. Sosto, V. Patti, P. Rosso, C. Bosco, J. Mata, E. Gualda, Multipride at evalita 2026: Overview of the multilingual automatic detection of slur reclamation in the lgbtq+ context task, in: Proceedings of the Ninth Evaluation Campaign of Natural Language Processin… find
  2. F. Cutugno, A. Miaschi, A. P. Aprosio, G. Rambelli, L. Siciliani, M. A. Stranisci, Evalita 2026: Overview of the 9th evaluation campaign of natural language processing and speech tools for italian, in: Proceedings of the Ninth Evaluation Campaign of Natural Language Processing and Speech Tools for I… find
  3. M. Junczys-Dowmunt, R. Grundkiewicz, T. Dwojak, H. Hoang, K. Heafield, T. Neckermann, et al., Marian: Fast neural machine translation in C++, in: Proceedings of ACL 2018, System Demonstrations, Melbourne, Australia, 2018, pp. 116–121. URL: https://aclanthology.org/P18-4020/. doi: 10.18653/v1/P18-402… — [doi:10.18653/v1/p18-4020](https://doi.org/10.18653/v1/p18-4020) ↗
  4. J. Tiedemann, M. Aulamo, D. Bakshandaeva, M. Boggia, S.-A. Grönroos, T. Nieminen, A. Raganato, Y. Scherrer, R. Vazquez, S. Virpioja, Democratizing neural machine translation with OPUS-MT, Language Resources and Evaluation (2023) 713–755. doi: 10.1007/s10579-023-09704-w . — [doi:10.1007/s10579-023-09704-w](https://doi.org/10.1007/s10579-023-09704-w) find
  5. J. Tiedemann, S. Thottingal, OPUS-MT — Building open translation services for the World, in: Proceedings of the 22nd Annual Conferenec of the European Association for Machine Translation (EAMT), Lisbon, Portugal, 2020. find
  6. J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional transformers for language understanding, in: Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short…
  7. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is all you need, Advances in neural information processing systems 30 (2017).
  8. L. Parisi, S. Francia, P. Magnani, Umberto: an italian language model trained with whole word masking, https://github.com/musixmatchresearch/umberto, 2020. ↗
  9. Y. Liu, M. Ott, N. Goyal, J. Du, M. Joshi, D. Chen, O. Levy, M. Lewis, L. Zettlemoyer, V. Stoyanov, Roberta: A robustly optimized bert pretraining approach, arXiv preprint arXiv:1907.11692 (2019).
  10. J. Cañete, G. Chaperon, R. Fuentes, J.-H. Ho, H. Kang, J. Pérez, Spanish pre-trained bert model and evaluation data, in: PML4DC at ICLR 2020, 2020. find
  11. A. Conneau, K. Khandelwal, N. Goyal, V. Chaudhary, G. Wenzek, F. Guzmán, et al., Unsupervised cross-lingual representation learning at scale, in: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020, pp. 8440–8451. find
  12. D. Nozza, Exposing the limits of zero-shot cross-lingual hate speech detection, in: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), Association for Computation…
  13. T.-Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollár, Focal loss for dense object detection, in: 2017 IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2999–3007. doi: 10.1109/ICCV. 2017.324 .
  14. I. Loshchilov, F. Hutter, Decoupled weight decay regularization, 2019. URL: https://arxiv.org/abs/ 1711.05101. arXiv:1711.05101 . ↗
  15. X. Chen, C. Liang, D. Huang, E. Real, K. Wang, H. Pham, X. Dong, T. Luong, C.-J. Hsieh, Y. Lu, Q. V. Le, Symbolic discovery of optimization algorithms, in: Proceedings of the 37th International Conference on Neural Information Processing Systems, Curran Associates Inc., Red Hook, NY, USA, 2023. find
  16. T. Akiba, S. Sano, T. Yanase, T. Ohta, M. Koyama, Optuna: A next-generation hyperparameter optimization framework, in: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, New York, NY, USA, 2019, p. 2623–2631. URL: https://doi.org/10.1145/3292500.333070… ↗
  17. A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, et al., Pytorch: An imperative style, high-performance deep learning library, in: Advances in Neural Information Processing Systems, volume 32, Curran Associates, Inc., 2019, pp. 8024–8035. URL: https://proceedings.neurips.cc/ paper/90… ↗
  18. T. Wolf, L. Debut, V. Sanh, J. Chaumond, C. Delangue, A. Moi, P. Cistac, T. Rault, R. Louf, M. Funtowicz, et al., Transformers: State-of-the-art natural language processing, in: Proceedings of the 2020 conference on empirical methods in natural language processing: system demonstrations, 2020, pp. 3…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/traina-2026-gruppettozzo-multipride-detecting-lgbtq-reclamatory/