⚡ 4 h early
last translation benchmark paper
Last Translation Benchmark
3 experts across 3 network communities independently surfaced this.
3 experts
3 communities
1 sources clustered
“Machine translation is not solved and it will take a while for it to be done arxiv.org/abs/2609.04173”
“Vil\'em Zouhar, Niyati Bafna, Mukund Choudhary, Maike Z\"ufle, Sara Rajaee, Pinzhen Chen, Jannis Vamvas, Sara Papi, Ona de Gibert, Bhavitvya Malik, Eliya Habba, Orfeas Menis Mastromichalakis, Patr\'icia Schmidtov\'a, ... Last Translation Benchmark https://a…”
2 experts discussed this · 5 posts
Vilém Zouhar: Machine translation is not solved and it will take a while for it to be done arxiv.org/abs/2609.04173
Vilém Zouhar: We just released the Last Translation Benchmark paper. In a massive crowdsourcing effort we collected 3456 unique hard-to-translate examples that break state-of-the-art translation models, and whic…
Vilém Zouhar: Machine translation doesn't break on just figurative language as one would expect. In fact, for the next generation of models, we may need to invest heavily into multilingual (& cultural) reasoning.
Open the full discussion →