{"type":"ScholarlyArticle","title":"Variational Learning for Unsupervised Knowledge Grounded Dialogs","authors":[{"name":"Mayank Mishra","slug":"mayank-mishra","given":"Mayank","family":"Mishra","affiliation":"IBM Research AI"},{"name":"Dhiraj Madan","slug":"dhiraj-madan","given":"Dhiraj","family":"Madan","affiliation":"IBM Research AI"},{"name":"Gaurav Pandey","slug":"gaurav-pandey","given":"Gaurav","family":"Pandey","affiliation":"IBM Research AI"},{"name":"Danish Contractor","slug":"danish-contractor","given":"Danish","family":"Contractor","affiliation":"IBM Research AI"}],"abstract":"Recent methods for knowledge grounded dialogs generate responses by incorporating information from an external textual document. These methods do not require the exact document to be known during training and rely on the use of a retrieval system to fetch relevant documents from a large index. The documents used to generate the responses are modeled as latent variables whose prior probabilities need to be estimated. Models such as RAG, marginalize the document probabilities over the documents retrieved from the index to define the log-likelihood loss function which is optimized end-to-end. \r\n\r\nIn this paper, we develop a variational approach to the above technique wherein, we instead maximize the Evidence Lower bound (ELBO). Using a collection of three publicly available open-conversation datasets, we demonstrate how the posterior distribution, which has information from the ground-truth response, allows for a better approximation of the objective function during training.  To overcome the challenges associated with sampling over a large knowledge collection, we develop an efficient approach to approximate the ELBO. \r\nTo the best of our knowledge, we are the first to apply variational training for open-scale unsupervised knowledge grounded dialog systems.","doi":"10.24963/ijcai.2022/597","url":"https://ijcai.org/proceedings/2022/597","pdf":"https://ijcai.org/proceedings/2022/0597.pdf","markdown":"https://ijcai.org/proceedings/2022/597.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/597","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Natural Language Processing","track":"Main Track","number":597,"pages":{"first":4303,"last":4309},"keywords":["Natural Language Processing: Language Grounding","Natural Language Processing: Dialogue and Interactive Systems"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986726","length_sec":92,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984869","length_sec":363,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Campos et al., 2020] Jon Ander Campos, Arantxa Otegi, Aitor Soroa, Jan Deriu, Mark Cieliebak, and Eneko Agirre. DoQA -accessing domain-specific FAQs via conversa- tional QA. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7302-7314, Online, July 2020. Association for Compu- tational Linguistics.","title":"DoQA -accessing domain-specific FAQs via conversational QA","authors":"Campos","year":"2020","venue":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chen et al., 2020] Xiuyi Chen, Fandong Meng, Peng Li, Feilong Chen, Shuang Xu, Bo Xu, and Jie Zhou. Bridging the gap between prior and posterior knowledge selection for knowledge-grounded dialogue generation. In Proceed- ings of the 2020 Conference on Empirical Methods in Nat- ural Language Processing (EMNLP), pages 3426-3437, 2020.","title":"Bridging the gap between prior and posterior knowledge selection for knowledge-grounded dialogue generation","authors":"Chen","year":"2020","venue":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Choi et al., 2018] Eunsol Choi, He He, Mohit Iyyer, Mark Yatskar, Wen-tau Yih, Yejin Choi, Percy Liang, and Luke Zettlemoyer. QuAC: Question answering in context. In Proceedings of the 2018 Conference on Empirical Meth- ods in Natural Language Processing, pages 2174-2184, Brussels, Belgium, October-November 2018. Association for Computational Linguistics.","title":"QuAC: Question answering in context","authors":"Choi","year":"2018","venue":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Devlin et al., 2018] Jacob Devlin, Ming-Wei Chang, Ken- ton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understand- ing. arXiv preprint arXiv:1810.04805, 2018.","title":"Bert: Pre-training of deep bidirectional transformers for language understanding","authors":"Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Guu et al., 2020] Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Ming-Wei Chang. Realm: Retrieval-augmented language model pre-training. arXiv preprint arXiv:2002.08909, 2020.","title":"Realm: Retrieval-augmented language model pre-training","authors":"Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, Ming-Wei Chang","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Jang et al., 2016] Eric Jang, Shixiang Gu, and Ben Poole. Categorical reparameterization with gumbel-softmax. arXiv preprint arXiv:1611.01144, 2016.","title":"Categorical reparameterization with gumbel-softmax","authors":"Eric Jang, Shixiang Gu, Ben Poole","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Johnson et al., 2017] Jeff Johnson, Matthijs Douze, and Hervé Jégou. Billion-scale similarity search with gpus. arXiv preprint arXiv:1702.08734, 2017.","title":"Billion-scale similarity search with gpus","authors":"Jeff Johnson, Matthijs Douze, Hervé Jégou","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Joshi et al., 2017] Mandar Joshi, Eunsol Choi, Daniel S Weld, and Luke Zettlemoyer. Triviaqa: A large scale distantly supervised challenge dataset for reading compre- hension. arXiv preprint arXiv:1705.03551, 2017.","title":"Triviaqa: A large scale distantly supervised challenge dataset for reading comprehension","authors":"Mandar Joshi, Eunsol Choi, Daniel S Weld, Luke Zettlemoyer","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Karpukhin et al., 2020] Vladimir Karpukhin, Barlas Oguz, Sewon Min, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. Dense passage retrieval for open-domain question answering. arXiv preprint arXiv:2004.04906, 2020.","title":"Dense passage retrieval for open-domain question answering","authors":"Vladimir Karpukhin, Barlas Oguz, Sewon Min, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-Tau Yih","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Kim et al., 2020a] Byeongchang Kim, Jaewoo Ahn, and Gunhee Kim. Sequential latent knowledge selec- tion for knowledge-grounded dialogue. arXiv preprint arXiv:2002.07510, 2020.","title":"Sequential latent knowledge selection for knowledge-grounded dialogue","authors":"Kim","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Kim et al., 2020b] Seokhwan Kim, Mihail Eric, Karthik Gopalakrishnan, Behnam Hedayatnia, Yang Liu, and Dilek Hakkani-Tur. Beyond domain APIs: Task-oriented conversational modeling with unstructured knowledge ac- cess. In Proceedings of the 21th Annual Meeting of SIG- DIAL, pages 278-289, 1st virtual meeting, July 2020. As- sociation for Computational Linguistics. [Kingma and Welling, 2013]","title":"Beyond domain APIs: Task-oriented conversational modeling with unstructured knowledge access","authors":"Kim","year":"2013","venue":"Proceedings of the 21th Annual Meeting of SIG-DIAL","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"Diederik P Kingma and Max Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.","title":"Auto-encoding variational bayes","authors":"P Diederik, Max Kingma, Welling","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Kwiatkowski et al., 2019] Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, et al. Natural questions: a benchmark for question answering research. Transactions of the Asso- ciation for Computational Linguistics, 7:453-466, 2019.","title":"Natural questions: a benchmark for question answering research","authors":"Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee","year":"2019","venue":"Transactions of the Association for Computational Linguistics","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Lewis et al., 2020] Patrick Lewis, Ethan Perez, Aleksan- dra Piktus, Fabio Petroni, Vladimir Karpukhin, Na- man Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, et al. Retrieval-augmented genera- tion for knowledge-intensive nlp tasks. arXiv preprint arXiv:2005.11401, 2020.","title":"Retrieval-augmented generation for knowledge-intensive nlp tasks","authors":"Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-Tau Yih, Tim Rocktäschel","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Li et al., 2020] Linxiao Li, Can Xu, Wei Wu, YUFAN ZHAO, Xueliang Zhao, and Chongyang Tao. Zero- resource knowledge-grounded dialogue generation. In H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, and H. Lin, editors, Advances in Neural Information Process- ing Systems, volume 33, pages 8475-8485.","title":"Zeroresource knowledge-grounded dialogue generation","authors":"Li","year":"2020","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Lian et al., 2019] Rongzhong Lian, Min Xie, Fan Wang, Jinhua Peng, and Hua Wu. Learning to select knowledge for response generation in dialog systems. arXiv preprint arXiv:1902.04911, 2019.","title":"Learning to select knowledge for response generation in dialog systems","authors":"Lian","year":"2019","venue":null,"doi":null,"url":null,"internal_url":"/proceedings/2019/706"},{"position":17,"raw":"[Loshchilov and Hutter, 2017] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017.","title":"Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization","authors":"Hutter Loshchilov","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Maddison et al., 2016] Chris J Maddison, Andriy Mnih, and Yee Whye Teh. The concrete distribution: A continuous relaxation of discrete random variables. arXiv preprint arXiv:1611.00712, 2016.","title":"The concrete distribution: A continuous relaxation of discrete random variables","authors":"Andriy Chris J Maddison, Yee Whye Mnih, Teh","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Paranjape et al., 2022] Ashwin Paranjape, Omar Khattab, Christopher Potts, Matei Zaharia, and Christopher D Man- ning. Hindsight: Posterior-guided training of retrievers for improved open-ended generation. In International Confer- ence on Learning Representations, 2022.","title":"Hindsight: Posterior-guided training of retrievers for improved open-ended generation","authors":"Ashwin Paranjape, Omar Khattab, Christopher Potts, Matei Zaharia, Christopher D Manning","year":"2022","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Qu et al., 2020] Chen Qu, Liu Yang, Cen Chen, Minghui Qiu, W. Bruce Croft, and Mohit Iyyer. Open-Retrieval Conversational Question Answering, page 539-548. As- sociation for Computing Machinery, New York, NY, USA, 2020.","title":"Open-Retrieval Conversational Question Answering","authors":"Chen Qu, Liu Yang, Cen Chen, Minghui Qiu, W Bruce Croft, Mohit Iyyer","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Radford et al., 2019] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019.","title":"Language models are unsupervised multitask learners","authors":"Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever","year":"2019","venue":"OpenAI blog","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Robertson et al., 1994] Stephen E. Robertson, Steve Walker, Susan Jones, Micheline Hancock-Beaulieu, and Mike Gatford. Okapi at trec-3. In TREC, 1994.","title":"Okapi at trec-3","authors":"Stephen E Robertson, Steve Walker, Susan Jones, Micheline Hancock-Beaulieu, Mike Gatford","year":"1994","venue":"TREC","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}