6 papers
Agreement-Constrained Probabilistic Minimum Bayes Risk Decoding
Koki Natsumi, Hiroyuki Deguchi, Yusuke Sakai +2
Minimum Bayes risk (MBR) decoding generates high-quality translations by maximizing the expected utility of output candidates, but it evaluates all pairwise scores over the candida…
Case-Based Decision-Theoretic Decoding with Quality Memories
Hiroyuki Deguchi, Masaaki Nagata
Minimum Bayes risk (MBR) decoding is a decision rule of text generation, which selects the hypothesis that maximizes the expected utility and robustly generates higher-quality text…
Diversity Explains Inference Scaling Laws: Through a Case Study of Minimum Bayes Risk Decoding
Hidetaka Kamigaito, Hiroyuki Deguchi, Yusuke Sakai +2
Inference methods play an important role in eliciting the performance of large language models (LLMs). Currently, LLMs use inference methods utilizing generated multiple samples, w…
Long-Tail Crisis in Nearest Neighbor Language Models
Yuto Nishida, Makoto Morishita, Hiroyuki Deguchi +2
The -nearest-neighbor language model (NN-LM), one of the retrieval-augmented language models, improves the perplexity for given text by directly accessing a large datastore b…
SoftMatcha: A Soft and Fast Pattern Matcher for Billion-Scale Corpus Searches
Hiroyuki Deguchi, Go Kamoda, Yusuke Matsushita +4
Researchers and practitioners in natural language processing and computational linguistics frequently observe and analyze the real language usage in large-scale corpora. For that p…
mbrs: A Library for Minimum Bayes Risk Decoding
Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito +1
Minimum Bayes risk (MBR) decoding is a decision rule of text generation tasks that outperforms conventional maximum a posterior (MAP) decoding using beam search by selecting high-q…