Tower v2: Unbabel-IST 2024 Submission for the General MT Shared Task
Ricardo Rei, Jose Pombal, Nuno M. Guerreiro, João Alves, Pedro Henrique Martins, Patrick Fernandes, Helena Wu, Tania Vaz, Duarte Alves, Amin Farajian, Sweta Agrawal, Antonio Farinhas, José G. C. De Souza, André Martins
Proceedings of the Ninth Conference on Machine Translation (WMT), 2024
In this work, we present TOWER-V2, an improved iteration of the state-of-the-art openweight TOWER models, and the backbone of our submission to the WMT24 General Translation shared task. TOWER-V2 introduces key improvements including expanded language coverage, enhanced data quality, and increased model capacity up to 70B parameters. Our final submission combines these advancements with quality-aware decoding strategies, selecting translations based on multiple translation quality signals. The resulting system demonstrates significant improvement over previous versions, outperforming closed commercial systems like GPT-4O, CLAUDE-SONNET-3.5, and DEEPL even at a smaller 7B scale.