Same evaluation, more tokens: On the effect of input length for machine translation evaluation using Large Language Models.

Tobias Domhan, Dawei Zhu

Browse the full EMNLP paper archive.