PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
就在這幾天微軟發佈了一款參數量多達170億的史上最大的自然語言生成模型 Turing-NLG，在多種語言模型基準上均實現了SOTA。 值得關注的是Turing-NLG在文本摘要上的表現，由於它已經非常善於理解文本了，因此不需要太多的配對數據就可以實現比已有模型更好的效果。 從Facebook的BART、Google的PEGASUS到今天Microsoft的Truing-NLG，越來越大的預訓練數據
相關文章
- 1. A Deep Reinforced Model for Abstractive Summarization
- 2. #Paper Reading# Abstractive Sentence Summarization with Attentive Recurrent Neural Networks
- 3. Fast Abstractive SUmmarization with Reinforce-Selected Sentence Rewriting
- 4. Deep Communicating Agents for Abstractive Summarization論文分享
- 5. PEGASUS
- 6. #Paper Reading# A Neural Attention Model for Abstractive Sentence Summarization
- 7. Self-Attention Guided Copy Mechanism for Abstractive Summarization---ACL 2020
- 8. 多標籤分類：Global Encoding for Abstractive Summarization
- 9. 【讀】seq2seq——（2）Abstractive Sentence Summarization with Attentive Recurrent Neural Networks
- 10. NLG簡要調研(data2text+(Abstractive) Summarization)
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
- • 算法總結-股票買賣
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. A Deep Reinforced Model for Abstractive Summarization
- 2. #Paper Reading# Abstractive Sentence Summarization with Attentive Recurrent Neural Networks
- 3. Fast Abstractive SUmmarization with Reinforce-Selected Sentence Rewriting
- 4. Deep Communicating Agents for Abstractive Summarization論文分享
- 5. PEGASUS
- 6. #Paper Reading# A Neural Attention Model for Abstractive Sentence Summarization
- 7. Self-Attention Guided Copy Mechanism for Abstractive Summarization---ACL 2020
- 8. 多標籤分類：Global Encoding for Abstractive Summarization
- 9. 【讀】seq2seq——（2）Abstractive Sentence Summarization with Attentive Recurrent Neural Networks
- 10. NLG簡要調研(data2text+(Abstractive) Summarization)