Applications 2 - Dialog (11/02/2021)
- What are the issues in dialog
- Dialog datasets and evaluation
- Typical dialog models
- Reference: MultiWOZ (Budzianowski et al. 2018)
- Reference: SGD (Rastogi et al. 2019)
- Reference: STAR (Mosig et al. 2020)
- Reference: Taskmaster-2 (Byrne et al. 2020)
- Reference: ABCD (Chen et al. 2021)
- Reference: Structured Fusion Networks (Mehri et al. 2019)
- Reference: SOLOIST (Peng et al. 2020)
- Reference: SimpleTOD (Hosseini-Asl et al. 2020)
- Reference: MarCO (Wang et al. 2020)
- Reference: DailyDialog (Li et al. 2017)
- Reference: PersonaChat (Zhang et al. 2018)
- Reference: Topical-Chat (Gopalakrishnan et al. 2019)
- Reference: Wizards of Wikipedia (Dinan et al. 2018)
- Reference: Empathetic Dialogs (Rashkin et al. 2019)
- Reference: Neural Conversational Model (Vinyals and Le. 2015)
- Reference: HRED (Serban et al. 2016)
- Reference: Diversity Promoting Objective (Li et al. 2016)
- Reference: Transfer-Transfo (Wolf et al. 2019)
- Reference: DialoGPT (Zhang et al. 2019)
- Reference: Meena (Adiwardana et al. 2020)
- Reference: PLATO-2 (Bao et al. 2021)
- Reference: DialoGLUE (Mehri et al. 2020)
- Reference: ConVEx (Henderson and Vulic. 2020)
- Reference: GenSF (Mehri and Eskenazi. 2021)
- Reference: TripPy (Heck et al. 2020)
- Reference: USR (Mehri and Eskenazi. 2020)
- Reference: GRADE (Huang et al. 2020)
- Reference: HolisticEval (Pang et al. 2020)
- Reference: DSTC6 (Hori and Hori. 2017)
- Reference: FED (Mehri and Eskenazi. 2020)
- Reference: DSTC9 (Gunasekara et al. 2021)
- Reference: DynaEval (Zhang et al. 2021)
- Reference: Comprehensive Assessment (Yeh et al. 2021)
Slides: Dialog Slides