Elamur Research⌕
☰
home / papers / 2021 / BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data

BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data

Paper · 2021 · Haoyu Song, Yan Wang, Kaiyan Zhang et al.

Summary

Persona-based chatbots should keep a stable identity, but fine-tuned models such as GPT2 still contradict their assigned persona. The authors attribute this to scarce training data. Persona-dense corpora like PersonaChat are costly to crowd-source, and large social-media corpora rarely contain personal content. Their solution splits the task into two sub-problems that each have plenty of data: generating a response, and judging whether a response is consistent with a persona.

The resulting model, BoB (BERT-over-BERT), is built from three BERT-initialized modules:

  • an encoder for the persona and query;
  • an auto-regressive decoder that drafts a rough response representation;
  • a bidirectional decoder that refines that draft using the persona alone, without the query.

Because the third module ignores the query, it can be trained on ordinary NLI corpora (SNLI, MNLI). An unlikelihood loss lowers the probability of hypotheses that contradict their premise. The authors test BoB in low-resource PersonaChat settings and on the persona-sparse PersonalDialog corpus. In both automatic and human evaluation, it beats strong baselines on most metrics, especially persona consistency. This shows that general inference data can make up for limited personalized dialogue data.

Key points

  • Architecture: a BERT encoder plus two BERT decoders, D1 for generation and D2 for consistency understanding. D2 is trained with unlikelihood on non-dialogue NLI data (SNLI, MNLI), with persona treated as premise and response as hypothesis.
  • Evaluation data: PersonaChat (persona-dense, including low-resource subsets) and PersonalDialog (persona-sparse, with random and biased test sets).
  • Full PersonaChat: BoB reaches PPL 7.8, ΔP 76.1 and Per.C. 0.60. Without the unlikelihood loss, ΔP falls to 7.8 and Per.C. to 0.48.
  • PersonalDialog, biased test set: BoB scores PPL 19.5 and Per.C. 0.15, against 38.7 and 0.13 for AttentionRouting. On the random test set, no method wins consistently.
  • Ablations: the unlikelihood objective drives consistency understanding, and D2 accounts for most of the perplexity drop. Code is available at github.com/songhaoyu/BoB.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Harbin Institute of TechnologyHarbin · China76 works · 27,918 citations

📚 Shelves 4 Computer Science · 9,086 works Artificial Intelligence · 8,889 works Arts and Humanities · 9,296 works Language and Linguistics · 9,206 works

🧭 Topics 1 Dialogue systems · 401 works

Cite

Cite as Haoyu Song, Yan Wang, Kaiyan Zhang, Wei-Nan Zhang and Ting Liu. 2021. BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.acl-long.14

@inproceedings{song2021bob,
  title = {{BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data}},
  author = {Haoyu Song, Yan Wang, Kaiyan Zhang, Wei-Nan Zhang and Ting Liu},
  booktitle = {Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)},
  year = {2021},
  doi = {10.18653/v1/2021.acl-long.14},
  url = {https://doi.org/10.18653/v1/2021.acl-long.14}
}

References

38 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2015. Neural machine translation by jointly learning to align and translate. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. our card
  2. Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning. 2015. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 632–642, Lisbon, Portugal. Association for Computati… find
  3. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… our card
  4. Emily Dinan, Varvara Logacheva, Valentin Malykh, Alexander Miller, Kurt Shuster, Jack Urbanek, Douwe Kiela, Arthur Szlam, Iulian Serban, Ryan Lowe, et al. 2019. The second conversational intelligence challenge (convai2). arXiv preprint arXiv:1902.00098. our card
  5. Sergey Golovanov, Rauf Kurbanov, Sergey Nikolenko, Kyryl Truskovskyi, Alexander Tselousov, and Thomas Wolf. 2019. Large-scale transfer learning for natural language generation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 6053–6058, Florence, Ital… find
  6. Irina Higgins, David Amos, David Pfau, Sébastien Racanière, Loı̈c Matthey, Danilo J. Rezende, and Alexander Lerchner. 2018. Towards a definition of disentangled representations. CoRR, abs/1812.02230. — [arXiv:1812.02230](https://arxiv.org/abs/1812.02230) find
  7. Lee Humphreys, Phillipa Gill, and Balachander Krishnamurthy. 2014. Twitter: a content analysis of personal information. Information, Communication & Society, 17(7):843–857. find
  8. Hyunjik Kim and Andriy Mnih. 2018. Disentangling by factorising. In International Conference on Machine Learning, pages 2649–2658. PMLR. find
  9. Diederik P Kingma and Max Welling. 2013. Autoencoding variational bayes. arXiv preprint arXiv:1312.6114. our card
  10. J Richard Landis and Gary G Koch. 1977. The measurement of observer agreement for categorical data. biometrics, pages 159–174. find
  11. Jiwei Li, Michel Galley, Chris Brockett, Jianfeng Gao, and Bill Dolan. 2016a. A diversity-promoting objective function for neural conversation models. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pa… our card
  12. Jiwei Li, Michel Galley, Chris Brockett, Georgios Spithourakis, Jianfeng Gao, and Bill Dolan. 2016b. A persona-based neural conversation model. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 994–1003, Berlin, Germany. Associa… our card
  13. Margaret Li, Stephen Roller, Ilia Kulikov, Sean Welleck, Y-Lan Boureau, Kyunghyun Cho, and Jason Weston. 2020. Don’t say that! making inconsistent dialogue unlikely with unlikelihood training. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4715– 472… our card
  14. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9. find
  15. Zhaojiang Lin, Zihan Liu, Genta Indra Winata, Samuel Cahyawijaya, Andrea Madotto, Yejin Bang, Etsuko Ishii, and Pascale Fung. 2020. Xpersona: Evaluating multilingual personalized chatbot. arXiv preprint arXiv:2003.07568. find
  16. Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M Smith, et al. 2020. Recipes for building an open-domain chatbot. arXiv preprint arXiv:2004.13637. our card
  17. Qian Liu, Yihong Chen, Bei Chen, Jian-Guang Lou, Zixuan Chen, Bin Zhou, and Dongmei Zhang. 2020. You impress me: Dialogue generation via mutual persona perception. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1417–1427, Online. Association for Com… our card
  18. Haoyu Song, Yan Wang, Wei-Nan Zhang, Xiaojiang Liu, and Ting Liu. 2020a. Generate, delete and rewrite: A three-stage framework for improving persona consistency of dialogue generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5821– 5831, Onlin… find
  19. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. find
  20. Haoyu Song, Yan Wang, Wei-Nan Zhang, Zhengyu Zhao, Ting Liu, and Xiaojiang Liu. 2020b. Profile consistency identification for open-domain dialogue agents. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6651–6662, Online. Association for Compu… find
  21. Francesco Locatello, Michael Tschannen, Stefan Bauer, Gunnar Rätsch, Bernhard Schölkopf, and Olivier Bachem. 2019. Disentangling factors of variations using few labels. In International Conference on Learning Representations. find
  22. Andrea Madotto, Zhaojiang Lin, Yejin Bang, and Pascale Fung. 2020. The adapter-bot: All-in-one controllable conversational model. arXiv preprint arXiv:2008.12579. — [arXiv:2008.12579](https://arxiv.org/abs/2008.12579) find
  23. Andrea Madotto, Zhaojiang Lin, Chien-Sheng Wu, and Pascale Fung. 2019. Personalizing dialogue agents via meta-learning. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5454–5459, Florence, Italy. Association for Computational Linguistics. our card
  24. Pierre-Emmanuel Mazaré, Samuel Humeau, Martin Raison, and Antoine Bordes. 2018. Training millions of personalized dialogue agents. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2775–2779, Brussels, Belgium. Association for Computational Linguistics… our card
  25. Mor Naaman, Jeffrey Boase, and Chih-Hui Lai. 2010. Is it really about me? message content in social awareness streams. In Proceedings of the 2010 ACM conference on Computer supported cooperative work, pages 189–192. find
  26. Qiao Qian, Minlie Huang, Haizhou Zhao, Jingfang Xu, and Xiaoyan Zhu. 2018. Assigning personality/profile to a chatting machine for coherent conversation generation. In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI-18, pages 4279–4285. Internationa… find
  27. Haoyu Song, Wei-Nan Zhang, Yiming Cui, Dong Wang, and Ting Liu. 2019. Exploiting persona information for diverse generation of conversational responses. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI-19, pages 5190–5196. International Joint Confe… find
  28. Yiping Song, Zequn Liu, Wei Bi, Rui Yan, and Ming Zhang. 2020c. Learning to customize model structures for few-shot dialogue generation tasks. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5832– 5841, Online. Association for Computational Linguisti… find
  29. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In NIPS. our card
  30. Sean Welleck, Ilia Kulikov, Stephen Roller, Emily Dinan, Kyunghyun Cho, and Jason Weston. 2019a. Neural text generation with unlikelihood training. In International Conference on Learning Representations. our card
  31. Sean Welleck, Jason Weston, Arthur Szlam, and Kyunghyun Cho. 2019b. Dialogue natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3731–3741, Florence, Italy. Association for Computational Linguistics. our card
  32. Adina Williams, Nikita Nangia, and Samuel Bowman. 2018. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Pa… find
  33. Thomas Wolf, Victor Sanh, Julien Chaumond, and Clement Delangue. 2019. Transfertransfo: A transfer learning approach for neural network based conversational agents. arXiv preprint arXiv:1901.08149. find
  34. Liang Xu, Hai Hu, Xuanwei Zhang, Lu Li, Chenjie Cao, Yudong Li, Yechen Xu, Kai Sun, Dian Yu, Cong Yu, Yin Tian, Qianqian Dong, Weitang Liu, Bo Shi, Yiming Cui, Junyi Li, Jun Zeng, Rongzhao Wang, Weijian Xie, Yanting Li, Yina Patterson, Zuoyu Tian, Yiwen Zhang, He Zhou, Shaoweihua Liu, Zhe Zhao, Qipe… our card
  35. Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. 2018. Personalizing dialogue agents: I have a dog, do you have pets too? In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2204– 2213, Melbo… our card
  36. Xueliang Zhao, Wei Wu, Chongyang Tao, Can Xu, Dongyan Zhao, and Rui Yan. 2019. Low-resource knowledge-grounded dialogue generation. In International Conference on Learning Representations. find
  37. Yinhe Zheng, Guanyi Chen, Minlie Huang, Song Liu, and Xuan Zhu. 2019. Personalized dialogue generation with diversified traits. ArXiv, abs/1901.09672. — [arXiv:1901.09672](https://arxiv.org/abs/1901.09672) find
  38. Yinhe Zheng, Rongsheng Zhang, Minlie Huang, and Xiaoxi Mao. 2020. A pre-training based personalized dialogue generation model with persona-sparse data. Proceedings of the AAAI Conference on Artificial Intelligence, 34(05):9693–9700. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/song-2021-bob-bert-over-bert-training/