[NLG] Pretraining for Conditional Generation with Pseudo Self Attention
摘要： 這篇文章的主要idea就是如何用現在已經有的語言模型，稍微引導一下生成基於某些條件的話語。如給個情感，生成一個帶情感的句子。作者在GPT2的結構上稍微修改了self-attention機制，命名爲Pseudo-Self，在decoder的時候引入外部控制，並且這樣做能夠儘量小的影響GPT2之前的訓練參數，達到的很好的效果。 模型： 作者主要對比了前面兩種工作，關於這兩種工作不做闡述
相關文章
- 1. [NLG] Zero-Shot Dialog Generation with Cross-Domain Latent Actions
- 2. 5.Conditional Generation by RNN & Attention(李宏毅)
- 3. 論文學習-ReCoSa: Detecting the Relevant Contexts with Self-Attention for Multi-turn Dialogue Generation
- 4. A Variational U-Net for Conditional Appearance and Shape Generation
- 5. Self-Attention GAN 中的 self-attention 機制
- 6. CVPR2020《Exploring Self-attention for Image Recognition》
- 7. Exploring Self-attention for Image Recognition稿
- 8. 【AAAI2019】Exploring Answer Stance Detection with Recurrent Conditional Attention
- 9. attention(show, attention and tell: neural image caption generation with visual attention)
- 10. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • RxJava操作符（七）Conditional and Boolean
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. [NLG] Zero-Shot Dialog Generation with Cross-Domain Latent Actions
- 2. 5.Conditional Generation by RNN & Attention(李宏毅)
- 3. 論文學習-ReCoSa: Detecting the Relevant Contexts with Self-Attention for Multi-turn Dialogue Generation
- 4. A Variational U-Net for Conditional Appearance and Shape Generation
- 5. Self-Attention GAN 中的 self-attention 機制
- 6. CVPR2020《Exploring Self-attention for Image Recognition》
- 7. Exploring Self-attention for Image Recognition稿
- 8. 【AAAI2019】Exploring Answer Stance Detection with Recurrent Conditional Attention
- 9. attention(show, attention and tell: neural image caption generation with visual attention)
- 10. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention