Challenges in Building Intelligent Open-domain Dialog Systems
By Minlie Huang, Xiaoyan Zhu, Jianfeng Gao. 2019
355 citations recorded by the source corpus.
There is a resurgent interest in developing intelligent open-domain dialog systems due to the availability of large amounts of conversational data and the recent progress on neural approaches to conversational AI [33]. Unlike traditional task-oriented bots, an open-domain dialog system aims to establish long-term connections with users by satisfying the human need for communication, affection, and social belonging. This article reviews the recent work on neural approaches that are devoted to addressing three challenges in developing such systems: semantics, consistency, and interactiveness. Semantics requires a dialog system to not only understand the content of the dialog but also identify users’ emotional and social needs during the conversation. Consistency requires the system to demonstrate a consistent personality to win users’ trust and gain their long-term confidence. Interactiveness refers to the system’s ability to generate interpersonal responses to achieve particular social goals such as entertainment and conforming. The studies we select to present in this survey are based on our unique views and are by no means complete. Nevertheless, we hope that the discussion will inspire new research in developing more intelligent open-domain dialog systems.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing, Dialogue & Conversational AI
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Jason Aaron Edward Weston
- Jianfeng Gao
- Mohit Bansal
- Yejin Choi
- Luke S. Zettlemoyer
- Percy Liang
- Noah A. Smith
- Y-Lan Boureau
In Computer Vision and Pattern Recognition
- Jason Aaron Edward Weston
- Jianfeng Gao
- Luc Van Gool
- Dacheng Tao
- Trevor Darrell
- Xiaogang Wang
- Mohit Bansal
- Jitendra Malik
In Natural Language Processing
- Jason Aaron Edward Weston
- Y-Lan Boureau
- Kurt Shuster
- Jianfeng Gao
- Emily Dinan
- Eric Michael Smith
- Yejin Choi
- Minlie Huang
In Dialogue & Conversational AI
- Jason Aaron Edward Weston
- Kurt Shuster
- Jianfeng Gao
- Jing Xu
- Mojtaba Komeili
- Eric Michael Smith
- Stephen Roller
- Michel Galley
Citations in the corpus
- Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions (2020, background): “For more general recent surveys, see Gao et al. (2019); Jurafsky and Martin (2019); Huang, Zhu, and Gao (2020).”
- BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage (2022, methodology): “A number of fine-tuning datasets are crowdsourced and publicly released for use by the research community (Serban et al., 2015; Huang et al., 2020), such as the ones we will use for training the BlenderBot 3 model in this work (see §3.”
- Survey of Hallucination in Natural Language Generation (2022, background): “In contrast, the tolerance is relatively higher in dialogue systems because the desired characteristics are not only faithfulness but also user engagement, especially in open-domain dialogue systems [73].”
- Internet-Augmented Dialogue Generation (2021, methodology): “Datasets such as pushshift.io Reddit (Baumgartner et al., 2020), PersonaChat (Zhang et al., 2018) or Empa-thetic Dialogues (Rashkin et al., 2019) (see Huang et al. (2020) for a review) are typically employed to train the weights of a Transformer encoder-decoder.”
- Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading (2019, methodology): “Prior work, e.g, (Ghazvininejad et al., 2018; Zhang et al., 2018a; Huang et al., 2019), uses grounding in the form of independent snippets of text: Foursquare tips and background information about a given speaker.”
- Dialogue in the Wild: Learning from a Deployed Role-Playing Game with Humans and Bots (2021, background): “A number of crowdsourced or scraped datasets have been developed to that end, including Daily Dialogue (Li et al., 2017), PersonaChat (Zhang et al., 2018), Empathetic Dialogues (Rashkin et al., 2019) and Wizard of Wikipedia (Dinan et al., 2019b); see Huang et al. (2020) for a review.”
- GPT (Generative Pre-Trained Transformer)— A Comprehensive Review on Enabling Technologies, Potential Applications, Emerging Challenges, and Future Directions (2023, background): “Customers can quickly receive answers to their inquiries about room availability, booking procedures, facilities, and more by incorporating the GPT model into a hotel’s website or mobile app [118].”
- Pre-Trained Language Models for Text Generation: A Survey (2022, background): “on open topics such as daily life, sports and entertainment [80], while the latter is focused on assisting users to complete specific tasks, such as hotel reservation and product purchase [220].”
- The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systems (2022, neutral): “Chatbots are a promising technology for providing humans with social support in open-ended, “chit chat” settings (Brandtzaeg and Følstad, 2017; Huang et al., 2020; Liu et al., 2021b) and in many other more structured domains (Gao et al.”
- Towards Emotional Support Dialog Systems (2021, background): “3471 tant for building more engaging dialog systems (Zhou et al., 2018; Li et al., 2017; Zhou and Wang, 2018; Huber et al., 2018; Huang et al., 2020).”
- Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models (2023, background): “Our term “persona” draws upon the linguistics notion of “persona” as more malleable and constructed-in-the-moment than “identity” (Podesva et al., 2015) and on the HCI use of “persona” as a model of a hypothetical individual (Cooper, 1999; Blomkvist, 2002; Jettmar and Nass, 2002; Muller and Carey, 2”
- Exploring Prompt-based Few-shot Learning for Grounded Dialog Generation (2021, background): “…works have greatly enhanced dialog models through grounding model-generated dialogs on various external information (Ghazvininejad et al., 2018; Huang et al., 2020), such as Wikipedia documents (Dinan et al., 2018), persona descriptions (Zhang et al., 2018) or emotional support strategies (Liu…”
- A Computational Approach to Understanding Empathy Expressed in Text-Based Mental Health Support (2020, background): “Recent work has also developed proof-of-concept prototypes, such as ClientBot (Huang et al., 2020), for training users in counseling.”