Optimizing the Pearson correlation between hypothesis likelihood and an external quality score during fine-tuning improves LLM translation quality and turns log-likelihood into a competitive reference-free quality estimator.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Calibrating Translation Decoding with Quality Estimation on LLMs
Optimizing the Pearson correlation between hypothesis likelihood and an external quality score during fine-tuning improves LLM translation quality and turns log-likelihood into a competitive reference-free quality estimator.