Summary of Smurfcat at Pan 2024 Textdetox: Alignment Of Multilingual Transformers For Text Detoxification, by Elisei Rykov et al.
SmurfCat at PAN 2024 TextDetox: Alignment of Multilingual Transformers for Text Detoxification
by Elisei Rykov, Konstantin Zaytsev, Ivan Anisimov, Alexandr Voronin
First submitted to arxiv on: 7 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a solution for the Multilingual Text Detoxification task in the PAN-2024 competition by the SmurfCat team. The authors utilize data augmentation through machine translation and filtering to collect an additional multilingual parallel dataset, which is then used to fine-tune various sequence-to-sequence models like mT0 and Aya on a text detoxification task. They apply the ORPO alignment technique to the final model, resulting in a state-of-the-art performance for Ukrainian and near-state-of-the-art results for other languages. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve a problem with bad language online by training special models that can understand many languages. The authors use machine translation to make more training data and then fine-tune their models on this new data. They achieve great results, especially in the Ukrainian language. This is important because it can help improve online safety. |
Keywords
» Artificial intelligence » Alignment » Data augmentation » Translation
