Latent Attribute Control for Story Generation

Yu Siou Tang, Chung Hsien Wu

研究成果: Conference contribution

摘要

Neural open-domain story generation aims to generate long and fluent text as human writing. Recent work attempts to generate stories in fine-grained controls such as plot-like structure and ending valence. Although those outputs comply with the rules of grammar, they generally have logical conflicts and a lack of long-range cohesion because of explicit controlling. In this study, we propose to capture challenging story representation using latent variable modeling for the storytelling model, and we align the encoder output with story latent embeddings. Our approach and baselines are all built on the pre-trained BART language model. Experimental results demonstrated that our model largely improved compared to strong baselines on human evaluation. Human evaluators favored our generated stories, and the results were more relevant to the prompt and more coherent than the baselines.

原文English
主出版物標題2021 International Conference on Asian Language Processing, IALP 2021
編輯Deyi Xiong, Ridong Jiang, Yanfeng Lu, Minghui Dong, Haizhou Li
發行者Institute of Electrical and Electronics Engineers Inc.
頁面148-153
頁數6
ISBN(電子)9781665483117
DOIs
出版狀態Published - 2021
事件2021 International Conference on Asian Language Processing, IALP 2021 - Singapore, Singapore
持續時間: 2021 12月 112021 12月 13

出版系列

名字2021 International Conference on Asian Language Processing, IALP 2021

Conference

Conference2021 International Conference on Asian Language Processing, IALP 2021
國家/地區Singapore
城市Singapore
期間21-12-1121-12-13

All Science Journal Classification (ASJC) codes

  • 人類學
  • 人工智慧
  • 電腦視覺和模式識別
  • 訊號處理

指紋

深入研究「Latent Attribute Control for Story Generation」主題。共同形成了獨特的指紋。

引用此