{"type":"ScholarlyArticle","title":"Curriculum Learning for Natural Answer Generation","authors":[{"name":"Cao Liu","slug":"cao-liu","given":"Cao","family":"Liu","affiliation":"National Laboratory of Pattern Recognition , Institute of Automation , Chinese Academy of Sciences , Beijing , 100190 , China; University of Chinese Academy of Sciences , Beijing , 100049 , China"},{"name":"Shizhu He","slug":"shizhu-he","given":"Shizhu","family":"He","affiliation":"National Laboratory of Pattern Recognition , Institute of Automation , Chinese Academy of Sciences , Beijing , 100190 , China"},{"name":"Kang Liu","slug":"kang-liu","given":"Kang","family":"Liu","affiliation":"National Laboratory of Pattern Recognition , Institute of Automation , Chinese Academy of Sciences , Beijing , 100190 , China; University of Chinese Academy of Sciences , Beijing , 100049 , China"},{"name":"Jun Zhao","slug":"jun-zhao","given":"Jun","family":"Zhao","affiliation":"National Laboratory of Pattern Recognition , Institute of Automation , Chinese Academy of Sciences , Beijing , 100190 , China; University of Chinese Academy of Sciences , Beijing , 100049 , China"}],"abstract":"By reason of being able to obtain natural language responses, natural answers are more favored in real-world Question Answering (QA) systems. Generative models learn to automatically generate natural answers from large-scale question answer pairs (QA-pairs). However, they are suffering from the uncontrollable and uneven quality of QA-pairs crawled from the Internet. To address this problem, we propose a curriculum learning based framework for natural answer generation (CL-NAG), which is able to take full advantage of the valuable learning data from a noisy and uneven-quality corpus. Specifically, we employ two practical measures to automatically measure the quality (complexity) of QA-pairs. Based on the measurements, CL-NAG firstly utilizes simple and low-quality QA-pairs to learn a basic model, and then gradually learns to produce better answers with richer contents and more complete syntaxes based on more complex and higher-quality QA-pairs. In this way, all valuable information in the noisy and uneven-quality corpus could be fully exploited. Experiments demonstrate that CL-NAG outperforms the state-of-the-arts, which increases 6.8% and 8.7% in the accuracy for simple and complex questions, respectively.","doi":"10.24963/ijcai.2018/587","url":"https://ijcai.org/proceedings/2018/587","pdf":"https://ijcai.org/proceedings/2018/0587.pdf","markdown":"https://ijcai.org/proceedings/2018/587.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/587","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Natural Language Processing","track":"Main track","number":587,"pages":{"first":4223,"last":4229},"keywords":["Natural Language Processing: Natural Language Processing","Natural Language Processing: Dialogue","Natural Language Processing: Question Answering","Natural Language Processing: Natural Language Generation"],"videos":[],"references":[{"position":1,"raw":"[Bahdanau et al., 2015] Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. Neural machine translation by jointly learning to align and translate. In Proceedings of ICLR, 2015.","title":"Neural machine translation by jointly learning to align and translate","authors":"Dzmitry Bahdanau, Kyunghyun Cho, Yoshua Bengio","year":"2015","venue":"Proceedings of ICLR","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bengio et al., 2009] Yoshua Bengio, Jérôme Louradour, Ronan Collobert, and Jason Weston. Curriculum learning. In Proceedings of ICML, pages 41-48. ACM, 2009.","title":"Curriculum learning","authors":"Yoshua Bengio, Jérôme Louradour, Ronan Collobert, Jason Weston","year":"2009","venue":"Proceedings of ICML","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Berant et al., 2013] Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. Semantic parsing on Freebase from question-answer pairs. In Proceedings of EMNLP, pages 1533-1544, 2013.","title":"Semantic parsing on Freebase from question-answer pairs","authors":"Jonathan Berant, Andrew Chou, Roy Frostig, Percy Liang","year":"2013","venue":"Proceedings of EMNLP","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Bordes et al., 2014a] Antoine Bordes, Sumit Chopra, and Jason Weston. Question answering with subgraph embed- dings. In Proceedings of EMNLP, pages 615-620, 2014.","title":"Question answering with subgraph embeddings","authors":"Bordes","year":"2014","venue":"Proceedings of EMNLP","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Bordes et al., 2014b] Antoine Bordes, Jason Weston, and Nicolas Usunier. Open question answering with weakly supervised embedding models. In Proceedings of ECML- PKDD, pages 165-180, 2014.","title":"Open question answering with weakly supervised embedding models","authors":"Bordes","year":"2014","venue":"Proceedings of ECML-PKDD","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Bordes et al., 2015] Antoine Bordes, Nicolas Usunier, Sumit Chopra, and Jason Weston. Large-scale simple question answering with memory networks. CoRR, abs/1506.02075, 2015.","title":"Large-scale simple question answering with memory networks","authors":"Antoine Bordes, Nicolas Usunier, Sumit Chopra, Jason Weston","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Cho et al., 2014] Kyunghyun Cho, Bart van Merrienboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. Learning phrase representations using rnn encoder-decoder for statistical machine translation. In Proceedings of EMNLP, pages 1724-1734, 2014.","title":"Learning phrase representations using rnn encoder-decoder for statistical machine translation","authors":"Cho","year":"2014","venue":"Proceedings of EMNLP","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Eric and Manning, 2017] Mihail Eric and Christopher D Manning. Key-value retrieval networks for task-oriented dialogue. In Proceedings of SIGDIAL, 2017.","title":"Key-value retrieval networks for task-oriented dialogue","authors":"Mihail Eric, Christopher D Manning","year":"2017","venue":"Proceedings of SIGDIAL","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Gu et al., 2016] Jiatao Gu, Zhengdong Lu, Hang Li, and Victor O.K. Li. Incorporating copying mechanism in sequence-to-sequence learning. In Proceedings of ACL, pages 1631-1640, 2016.","title":"Incorporating copying mechanism in sequence-to-sequence learning","authors":"Jiatao Gu, Zhengdong Lu, Hang Li, O K Victor, Li","year":"2016","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Hao et al., 2017] Yanchao Hao, Yuanzhe Zhang, Kang Li- u, Shizhu He, Zhanyi Liu, Hua Wu, and Jun Zhao. An end-to-end model for question answering over knowledge base with cross-attention combining global knowledge. In Proceedings of ACL, pages 221-231, 2017.","title":"An end-to-end model for question answering over knowledge base with cross-attention combining global knowledge","authors":"Hao","year":"2017","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[He et al., 2017] Shizhu He, Cao Liu, Kang Liu, and Jun Zhao. Generating natural answers by incorporating copy- ing and retrieving mechanisms in sequence-to-sequence learning. In Proceedings of ACL, pages 199-208, 2017.","title":"Generating natural answers by incorporating copying and retrieving mechanisms in sequence-to-sequence learning","authors":"He","year":"2017","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Hochreiter and Schmidhuber, 1997] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735-1780, 1997.","title":"Long short-term memory","authors":"Schmidhuber Hochreiter, Sepp Hochreiter, Jürgen Schmidhuber","year":"1997","venue":"Neural computation","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Kumar et al., 2010] M Pawan Kumar, Benjamin Packer, and Daphne Koller. Self-paced learning for latent variable models. In Proceedings of NIPS, pages 1189-1197, 2010.","title":"Self-paced learning for latent variable models","authors":"Benjamin M Pawan Kumar, Daphne Packer, Koller","year":"2010","venue":"Proceedings of NIPS","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Levy and Manning, 2003] Roger Levy and Christopher D. Manning. Is it harder to parse chinese, or the chinese tree- bank? In Proceedings of ACL, pages 439-446, 2003.","title":"Is it harder to parse chinese, or the chinese treebank?","authors":"Roger Levy, Christopher D Manning","year":"2003","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Li et al., 2016] Jiwei Li, Will Monroe, Alan Ritter, Dan Ju- rafsky, Michel Galley, and Jianfeng Gao. Deep reinforce- ment learning for dialogue generation. In Proceedings of EMNLP, pages 1192-1202, 2016.","title":"Deep reinforcement learning for dialogue generation","authors":"Jiwei Li, Will Monroe, Alan Ritter, Dan Jurafsky, Michel Galley, Jianfeng Gao","year":"2016","venue":"Proceedings of EMNLP","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Mou et al., 2016] Lili Mou, Yiping Song, Rui Yan, Ge Li, Lu Zhang, and Zhi Jin. Sequence to backward and forward sequences: A content-introducing approach to generative short-text conversation. In Proceedings of COLING, pages 3349-3358, 2016.","title":"Sequence to backward and forward sequences: A content-introducing approach to generative short-text conversation","authors":"Lili Mou, Yiping Song, Rui Yan, Ge Li, Lu Zhang, Zhi Jin","year":"2016","venue":"Proceedings of COLING","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Novikova et al., 2017] Jekaterina Novikova, Ondej Duek, Amanda Cercas Curry, and Verena Rieser. Why we need new evaluation metrics for nlg. In Proceedings of EMNLP, pages 2241-2252, 2017.","title":"Why we need new evaluation metrics for nlg","authors":"Jekaterina Novikova, Ondej Duek, Amanda Cercas Curry, Verena Rieser","year":"2017","venue":"Proceedings of EMNLP","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Ranzato et al., 2015] Marc'Aurelio Ranzato, Sumit Chopra, Michael Auli, and Wojciech Zaremba. Sequence level training with recurrent neural networks. Proceedings of ICLR, 2015.","title":"Sequence level training with recurrent neural networks","authors":"Ranzato","year":"2015","venue":"Proceedings of ICLR","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Sachan and Xing, 2016] Mrinmaya Sachan and Eric Xing. Easy questions first? a case study on curriculum learn- ing for question answering. In Proceedings of ACL, pages 453-463, 2016.","title":"Easy questions first? a case study on curriculum learning for question answering","authors":"Xing Sachan, Mrinmaya Sachan, Eric Xing","year":"2016","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Salton and McGill, 1986] Gerard Salton and Michael J M- cGill. Introduction to modern information retrieval. 1986.","title":"Introduction to modern information retrieval","authors":"Gerard Salton, J M-Cgill Michael","year":"1986","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Sharma et al., 2017] Shikhar Sharma, Layla El Asri, Hannes Schulz, and Jeremie Zumer. Relevance of unsu- pervised metrics in task-oriented dialogue for evaluating natural language generation. CoRR, abs/1706.09799, 2017.","title":"Relevance of unsupervised metrics in task-oriented dialogue for evaluating natural language generation","authors":"Shikhar Sharma, Layla El Asri, Hannes Schulz, Jeremie Zumer","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Skinner, 1958] Burrhus F Skinner. Reinforcement today. American Psychologist, 13(3):94, 1958.","title":"Reinforcement today","authors":"F Burrhus, Skinner","year":"1958","venue":"American Psychologist","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Sutskever et al., 2014] Ilya Sutskever, Oriol Vinyals, and Quoc V Le. Sequence to sequence learning with neu- ral networks. In Proceedings of NIPS, pages 3104-3112, 2014.","title":"Quoc V Le. Sequence to sequence learning with neural networks","authors":"Sutskever","year":"2014","venue":"Proceedings of NIPS","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Unger et al., 2014] Christina Unger, André Freitas, and Philipp Cimiano. An introduction to question answering over linked data. In Reasoning Web. Reasoning on the We- b in the Big Data Era, pages 100-140. Springer, 2014.","title":"An introduction to question answering over linked data","authors":"Christina Unger, André Freitas, Philipp Cimiano","year":"2014","venue":"Reasoning Web. Reasoning on the Web in the Big Data Era","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yao and Van Durme, 2014] Xuchen Yao and Benjamin Van Durme. Information extraction over structured data: Question answering with freebase. In Proceedings of ACL, pages 956-966, 2014.","title":"Information extraction over structured data: Question answering with freebase","authors":"Van Yao, Xuchen Durme, Benjamin Yao, Van Durme","year":"2014","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Yih et al., 2016] Wen-tau Yih, Matthew Richardson, Chris Meek, Ming-Wei Chang, and Jina Suh. The value of se- mantic parse labeling for knowledge base question answer- ing. In Proceedings of ACL, pages 201-206, 2016.","title":"The value of semantic parse labeling for knowledge base question answering","authors":"Wen-Tau Yih, Matthew Richardson, Chris Meek, Ming-Wei Chang, Jina Suh","year":"2016","venue":"Proceedings of ACL","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Yin et al., 2016] Jun Yin, Xin Jiang, Zhengdong Lu, Lifeng Shang, Hang Li, and Xiaoming Li. Neural generative question answering. In Proceedings of IJCAI, 2016.","title":"Neural generative question answering","authors":"Jun Yin, Xin Jiang, Zhengdong Lu, Lifeng Shang, Hang Li, Xiaoming Li","year":"2016","venue":"Proceedings of IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2016/422"}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}