Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.12677 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909790521786368 |
|---|---|
| author | Deguchi, Hiroyuki Nagata, Masaaki |
| author_facet | Deguchi, Hiroyuki Nagata, Masaaki |
| contents | 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 texts than maximum a posteriori (MAP) decoding. However, it depends on sample texts drawn from the text generation model; thus, it is difficult to find a hypothesis that correctly captures the knowledge or information of out-of-domain. To tackle this issue, we propose case-based decision-theoretic (CBDT) decoding, another method to estimate the expected utility using examples of domain data. CBDT decoding not only generates higher-quality texts than MAP decoding, but also the combination of MBR and CBDT decoding outperformed MBR decoding in seven domain De--En and Ja$\leftrightarrow$En translation tasks and image captioning tasks on MSCOCO and nocaps datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_12677 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Case-Based Decision-Theoretic Decoding with Quality Memories Deguchi, Hiroyuki Nagata, Masaaki Computation and Language 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 texts than maximum a posteriori (MAP) decoding. However, it depends on sample texts drawn from the text generation model; thus, it is difficult to find a hypothesis that correctly captures the knowledge or information of out-of-domain. To tackle this issue, we propose case-based decision-theoretic (CBDT) decoding, another method to estimate the expected utility using examples of domain data. CBDT decoding not only generates higher-quality texts than MAP decoding, but also the combination of MBR and CBDT decoding outperformed MBR decoding in seven domain De--En and Ja$\leftrightarrow$En translation tasks and image captioning tasks on MSCOCO and nocaps datasets. |
| title | Case-Based Decision-Theoretic Decoding with Quality Memories |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.12677 |