Improving Open-Ended Text Generation via Adaptive Decoding
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910468732354560 |
|---|---|
| author | Zhu, Wenhong Hao, Hongkun He, Zhiwei Ai, Yiming Wang, Rui |
| author_facet | Zhu, Wenhong Hao, Hongkun He, Zhiwei Ai, Yiming Wang, Rui |
| contents | Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers language models to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of language models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_18223 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Improving Open-Ended Text Generation via Adaptive Decoding Zhu, Wenhong Hao, Hongkun He, Zhiwei Ai, Yiming Wang, Rui Computation and Language Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers language models to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of language models. |
| title | Improving Open-Ended Text Generation via Adaptive Decoding |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2402.18223 |