Noisy-Channel Minimum Bayes Risk Decoding

Noisy-channel decomposition of Minimum Bayes Risk decoding addresses asymmetric metric bias

A new arXiv preprint proposes a noisy-channel decomposition of Minimum Bayes Risk (MBR) decoding that breaks the process into four components: hypothesis-to-reference likelihood, reference-to-hypothesis likelihood, hypothesis prior, and reference prior. The decomposition addresses a known asymmetry problem in MBR decoding caused by directional evaluation metrics like BLEU and COMET. The framework unifies existing MBR variants under a single interpretation and suggests that channel-specific weighting could improve over standard MBR decoding.