Beyond Prompts: A Systematic Study of LLM-Based Machine Translation Evaluators
LLM-based evaluators have rapidly become a standard tool for machine translation evaluation, yet the design choices that shape their behavior remain poorly understood. We present a systematic study of how instruction design, few-shot demonstrations, backbone model choice, and multi-step workflows affect span-level and...
Dmitry Popov, Roman Bokhyan, Ekaterina V. Enikeeva et al.
· Conference on Empirical Meth... · 0 citations