Elamur Research⌕
☰
home / papers / 2019 / MoEL: Mixture of Empathetic Listeners

MoEL: Mixture of Empathetic Listeners

Summary

Most earlier empathetic chatbots either produce text in a predefined emotion or train one decoder that predicts emotion and generates replies together. Neither approach models how a reply should change depending on what the user feels. MoEL (Mixture of Empathetic Listeners) addresses this with an end-to-end design. A transformer encoder serves as an emotion tracker: it reads the flattened multi-turn context and outputs a probability distribution over possible user emotions. Each emotion has its own transformer decoder, called a "listener", and a shared listener captures general context. A meta listener then mixes the listeners' output states, weighting each by the predicted emotion probabilities, and generates the response. The authors tested MoEL on the empatheticdialogues corpus (Rashkin et al., 2018) against a plain Transformer and a multitask baseline. Human judges preferred MoEL for empathy and relevance, and the abstract also reports gains in fluency. Because individual listeners can be inspected, the model's generation process is more interpretable. This shows that structuring the decoder around emotions can make dialogue agents both more empathetic and easier to analyse.

Key points

  • Architecture: MoEL has three parts: an emotion tracker (which also encodes the context), emotion-aware listeners, and a meta listener.
    • A BERT-style QRY token is prepended to the input, and its encoding serves as the query for emotion classification.
    • Inputs combine word, positional, and dialogue-state embeddings so the model can tell speaker turns from listener turns.
  • Weighting: Listener weights come from a Key-Value Memory Network, using a dot product followed by a softmax.
    • These weights are supervised with a cross-entropy loss on the gold speaker emotion.
    • The shared listener always receives a fixed weight of 1.
  • Evaluation: Experiments use empatheticdialogues, an open-domain, multi-turn corpus grounded in emotional situations. Human evaluation covered empathy, relevance, and fluency.
  • Analysis:
    • Heatmaps of the emotion distribution show the model can blend several listeners, for example combining annoyed and afraid.
    • It sometimes fixates on misleading cues, such as the word "surprised".
    • A listener whose emotion clashes with the actual context produces replies that are emotionally inappropriate.
  • Code: Released at github.com/HLTCHKUST/MoEL. Suggested future directions are adding personas and integrating with task-oriented dialogue systems.

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

Fields: Natural language processing

Cited by

88 papers of our library cite it (of the 140884 whose bibliographies we read); 7 of them have a page here, 81 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/d19-1012

@inproceedings{lin2019moel,
  title = {{MoEL: Mixture of Empathetic Listeners}},
  author = {Lin, Zhaojiang and Madotto, Andrea and Shin, Jamin and Xu, Peng and Fung, Pascale},
  booktitle = {Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)},
  year = {2019},
  doi = {10.18653/v1/d19-1012},
  url = {https://doi.org/10.18653/v1/d19-1012}
}

References

56 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Yingruo Fan, Jacqueline CK Lam, and Victor OK Li. 2018b. Unsupervised domain adaptation with generative adversarial networks for facial emotion recognition. In 2018 IEEE International Conference on Big Data (Big Data), pages 4460–4464. IEEE. find
  2. Rahaf Aljundi, Punarjay Chakravarty, and Tinne Tuytelaars. 2017. Expert gate: Lifelong learning with a network of experts. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 3366–3375. find
  3. Yingruo Fan, Jacqueline CK Lam, and Victor OK Li. 2018c. Video-based emotion recognition using deeply-supervised neural networks. In Proceedings of the 2018 on International Conference on Multimodal Interaction, pages 584–588. ACM. find
  4. Dario Bertero, Farhad Bin Siddique, Chien-Sheng Wu, Yan Wan, Ricky Ho Yin Chan, and Pascale Fung. 2016. Real-time speech emotion and sentiment Jiwei Li, Will Monroe, Alan Ritter, Dan Jurafsky, Michel Galley, and Jianfeng Gao. 2016c. Deep reinforcement learning for dialogue generation. In Proceedings… find our card
  5. Jianfeng Gao, Michel Galley, and Lihong Li. 2018. Neural approaches to conversational ai. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, pages 1371–1374. ACM. find
  6. Braden Hancock, Antoine Bordes, Pierre-Emmanuel Mazare, and Jason Weston. 2019. Learning from dialogue after deployment: Feed yourself, chatbot! arXiv preprint arXiv:1901.05415. find
  7. Chia-Wei Liu, Ryan Lowe, Iulian Serban, Mike Noseworthy, Laurent Charlin, and Joelle Pineau. 2016. How not to evaluate your dialogue system: An empirical study of unsupervised evaluation metrics for dialogue response generation. In Proceedings of the 2016 Conference on Empirical Methods in Natural L… find
  8. Zhiting Hu, Zichao Yang, Xiaodan Liang, Ruslan Salakhutdinov, and Eric P Xing. 2017. Toward controlled generation of text. In International Conference on Machine Learning, pages 1587–1596. find our card
  9. Nurul Lubis, Sakriani Sakti, Koichiro Yoshino, and Satoshi Nakamura. 2018. Eliciting positive emotion through affect-sensitive dialogue response generation: A neural network approach. In Thirty-Second AAAI Conference on Artificial Intelligence. find
  10. Robert A Jacobs, Michael I Jordan, Steven J Nowlan, Geoffrey E Hinton, et al. 1991. Adaptive mixtures of local experts. Neural computation, 3(1):79–87. find
  11. Andrea Madotto, Zhaojiang Lin, Chien-Sheng Wu, and Pascale Fung. 2019. Personalizing dialogue agents via meta-learning. In Proceedings of the 57th Conference of the Association for Computational Linguistics, pages 5454–5459. find
  12. Michael I Jordan and Robert A Jacobs. 1994. Hierarchical mixtures of experts and the em algorithm. Neural computation, 6(2):181–214. find
  13. Chaitanya K Joshi, Fei Mi, and Boi Faltings. 2017. Personalization in goal-oriented dialog. arXiv preprint arXiv:1706.07503. — [arXiv:1706.07503](https://arxiv.org/abs/1706.07503) find
  14. Andrea Madotto, Chien-Sheng Wu, and Pascale Fung. 2018. Mem2seq: Effectively incorporating knowledge bases into end-to-end task-oriented dialog systems. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1468–1478. find our card
  15. Lukasz Kaiser, Aidan N Gomez, Noam Shazeer, Ashish Vaswani, Niki Parmar, Llion Jones, and Jakob Uszkoreit. 2017. One model to learn them all. arXiv preprint arXiv:1706.05137. — [arXiv:1706.05137](https://arxiv.org/abs/1706.05137) find
  16. Pierre-Emmanuel Mazare, Samuel Humeau, Martin Raison, and Antoine Bordes. 2018a. Training millions of personalized dialogue agents. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2775–2779. find our card
  17. Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980. find
  18. Ilya Kulikov, Alexander H Miller, Kyunghyun Cho, and Jason Weston. 2018. Importance of a search strategy in neural dialogue modelling. arXiv preprint arXiv:1811.00907. — [arXiv:1811.00907](https://arxiv.org/abs/1811.00907) find
  19. Pierre-Emmanuel Mazare, Samuel Humeau, Martin Raison, and Antoine Bordes. 2018b. Training millions of personalized dialogue agents. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2775–2779. Association for Computational Linguistics. find our card
  20. Nayeon Lee, Zihan Liu, and Pascale Fung. 2019. Team yeon-zi at semeval-2019 task 4: Hyperpartisan news detection by de-noising weakly-labeled data. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 1052–1056. find
  21. Alexander Miller, Adam Fisch, Jesse Dodge, Amir-Hossein Karimi, Antoine Bordes, and Jason Weston. 2016. Key-value memory networks for directly reading documents. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 1400–1409. find
  22. 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… find our card
  23. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting on association for computational linguistics, pages 311–318. Association for Computational Linguistics. find our card
  24. 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), volume 1, pages 994–1003. find our card
  25. Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pages 1532–1543. find
  26. Ke Wang and Xiaojun Wan. 2018. Sentigan: generating sentimental texts via mixture adversarial networks. In Proceedings of the 27th International Joint Conference on Artificial Intelligence, pages 4446–4452. AAAI Press. find
  27. Dinesh Raghu, Nikhil Gupta, and Mausam. 2019. Disentangling Language and Knowledge in Task-Oriented Dialogs. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 1239… find
  28. Jason Weston, Emily Dinan, and Alexander Miller. 2018. Retrieve and refine: Improved sequence generation models for dialogue. In Proceedings of the 2018 EMNLP Workshop SCAI: The 2nd International Workshop on Search-Oriented Conversational AI, pages 87–92. find our card
  29. Hannah Rashkin, Eric Michael Smith, Margaret Li, and Y-Lan Boureau. 2018. I know the feeling: Learning to converse with empathy. arXiv preprint arXiv:1811.00207. find
  30. Genta Indra Winata, Onno Kampman, Yang Yang, Anik Dey, and Pascale Fung. 2017. Nora the empathetic psychologist. find
  31. Proc. Interspeech 2017, pages 3437–3438. find
  32. Carl E Rasmussen and Zoubin Ghahramani. 2002. Infinite mixtures of gaussian process experts. In Advances in neural information processing systems, pages 881–888. find
  33. Genta Indra Winata, Andrea Madotto, Zhaojiang Lin, Jamin Shin, Yan Xu, Peng Xu, and Pascale Fung. 2019. Caire hkust at semeval-2019 task 3: Hierarchical attention for dialogue emotion classification. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 142–147. find
  34. Revanth Reddy, Danish Contractor, Dinesh Raghu, and Sachindra Joshi. 2018. Multi-level memory for task oriented dialogs. arXiv preprint arXiv:1810.10647. find
  35. Jurgen Schmidhuber. 1987. Evolutionary principles in self-referential learning. on learning now to learn: The meta-meta-meta...-hook. Diploma thesis, Technische Universitat Munchen, Germany, 14 May. find
  36. 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. — [arXiv:1901.08149](https://arxiv.org/abs/1901.08149) find
  37. Iulian Vlad Serban, Ryan Lowe, Laurent Charlin, and Joelle Pineau. 2016. Generative deep neural networks for dialogue: A short review. arXiv preprint arXiv:1611.06216. — [arXiv:1611.06216](https://arxiv.org/abs/1611.06216) find
  38. Chien-Sheng Wu, Andrea Madotto, Genta Winata, and Pascale Fung. 2017. End-to-end recurrent entity network for entity-value independent goal-oriented dialog learning. In Dialog System Technology Challenges Workshop, DSTC6. find
  39. Babak Shahbaba and Radford Neal. 2009. Nonlinear models using dirichlet process mixtures. Journal of Machine Learning Research, 10(Aug):1829–1850. find
  40. Chien-Sheng Wu, Andrea Madotto, Genta Indra Winata, and Pascale Fung. 2018a. End-to-end dynamic query memory network for entity-value independent task-oriented dialog. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6154–6158. IEEE. find
  41. Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz, Andy Davis, Quoc Le, Geoffrey Hinton, and Jeff Dean. 2017. Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. arXiv preprint arXiv:1701.06538. find
  42. Jamin Shin, Peng Xu, Andrea Madotto, and Pascale Fung. 2019. Happybot: Generating empathetic dialogue responses by improving user experience lookahead. arXiv preprint arXiv:1906.08487. — [arXiv:1906.08487](https://arxiv.org/abs/1906.08487) find
  43. Chien-Sheng Wu, Richard Socher, and Caiming Xiong. 2019. Global-to-local memory pointer networks for task-oriented dialogue. arXiv preprint arXiv:1901.04713. — [arXiv:1901.04713](https://arxiv.org/abs/1901.04713) find
  44. Lucas Theis and Matthias Bethge. 2015. Generative image modeling using spatial lstms. In Advances in Neural Information Processing Systems, pages 1927–1935. find
  45. Yu Wu, Furu Wei, Shaohan Huang, Zhoujun Li, and Ming Zhou. 2018b. Response generation by context-aware prototype editing. arXiv preprint arXiv:1806.07042. — [arXiv:1806.07042](https://arxiv.org/abs/1806.07042) find
  46. Volker Tresp. 2001. Mixtures of gaussian processes. In Advances in neural information processing systems, pages 654–660. find
  47. Peng Xu, Andrea Madotto, Chien-Sheng Wu, Ji Ho Park, and Pascale Fung. 2018. Emo2vec: Learning generalized emotion representation by multi-task training. In Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, pages 292–298. find
  48. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Advances in neural information processing systems, pages 5998–6008. find
  49. Bangpeng Yao, Dirk Walther, Diane Beck, and Li Fei-Fei. 2009. Hierarchical mixture of classification experts uncovers interactions between brain regions. In Advances in Neural Information Processing Systems, pages 2178–2186. find
  50. Oriol Vinyals and Quoc Le. 2015. A neural conversational model. In International Conference on Machine Learning. find our card
  51. Semih Yavuz, Abhinav Rastogi, Guanlin Chao, Dilek Hakkani-Tür, and Amazon Alexa AI. 2018. Deepcopy: Grounded response generation with hierarchical pointer networks. ConvAI Workshop@NIPS. find
  52. Yury Zemlyanskiy and Fei Sha. 2018. Aiming to know you better perhaps makes me a more engaging dialogue partner. find
  53. Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. 2018a. 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. find our card
  54. Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. 2018b. 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. Asso… find our card
  55. Hao Zhou, Minlie Huang, Tianyang Zhang, Xiaoyan Zhu, and Bing Liu. 2018. Emotional chatting machine: Emotional conversation generation with internal and external memory. In Thirty-Second AAAI Conference on Artificial Intelligence. find
  56. Xianda Zhou and William Yang Wang. 2018. Mojitalk: Generating emotional responses at scale. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1128–1137. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/lin-2019-moel-mixture-empathetic-listeners/