Alana: Social Dialogue using an Ensemble Model and a Ranker trained on User Feedback
By Ioannis V. Papaioannou, Amanda Cercas Curry, Jose L. Part, Igor Shalyminov, Xu Xinnuo, Yanchao Yu, Ondrej Dusek, Verena Rieser, Oliver Lemon. 2017
45 citations recorded by the source corpus.
Subfields: Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Natural Language Processing
- Emily Dinan
- Y-Lan Boureau
- Jason Aaron Edward Weston
- Verena Rieser
- Gavin Abercrombie
- Dirk Hovy
- Shannon L. Spruit
- Dilek Z. Hakkani-Tür
Citations in the corpus
- Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling (2021, background): “1 (Papaioannou et al., 2017) reading out a news article by the Washington post.”
- Towards Coherent and Engaging Spoken Dialog Response Generation Using Automatic Conversation Evaluators (2019, methodology): “There has been recent work which combines responses from multiple sub-systems (Serban et al., 2017; Papaioannou et al., 2017) and ranks them to output the final sys-tem response.”
- SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems (2022, background): “ALANA v.1 (Papaioannou et al., 2017) reading out part of a news article (Wootson, 2017).”
- Beyond User Self-Reported Likert Scale Ratings: A Comparison Model for Automatic Dialog Evaluation (2020, background): “One of the Alexa Price Challenge team, Alana (Papaioannou et al., 2017) train a binaryclassifier between successful dialogs (human rating 4 or 5) and unsuccessful dialogs (rating 1 or 2) with heavy hand-engineered features.”
- Spoken Conversational AI in Video Games: Emotional Dialogue Management Increases User Engagement (2018, methodology): “It then sends the responses from those services to a lightweight version of the open-domain social conversational AI system Alana presented in [11, 12].”
- What Makes a Good Conversation?: Challenges in Designing Truly Conversational Agents (2019, background): “ions that ‘conversation’ is a poor description of the current interaction experience [37, 42]. This deficit has motivated a number of efforts to improve the conversational capabilities in such agents [26, 41]. Recent work has looked to imbue systems with the ability to engage in social talk [52] and”