NLP seminar: human-computer interaction 2021/22 fall

Topics

1. Recipes for building an open-domain chatbot (September 24, 2021)
   • Bonus paper: Multi-Modal Open-Domain Dialogue
   • This was a long paper, so it counts as 2 summaries and 2 discussions
2. Designing Precise and Robust Dialogue Response Evaluators (October 1, 2021)
3. Target-Guided Open-Domain Conversation (October 8, 2021)
4. Modeling Long Context for Task-Oriented Dialogue State Generation (October 15, 2021)
   Submit your review here
5. Towards the Use of Pretrained Language Models for Task-Oriented Dialogue Systems (October 22, 2021)
   Submit your review here
6. Multitask Prompted Training Enables Zero-Shot Task Generalization (October 29, 2021)
   Submit your review here
7. Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue (November 5, 2021)
   Submit your review here
8. Towards Incremental Transformers: An Empirical Analysis of Transformer Models for Incremental NLU (November 12, 2021)
   Submit your review here
9. Surface Form Competition: Why the Highest Probability Answer Isn’t Always Right (November 19, 2021)
   Submit your review here
10. Few-Shot Intent Detection via Contrastive Pre-Training and Fine-Tuning (November 26, 2021)
    Submit your review here
11. End-to-End Learning of Flowchart Grounded Task-Oriented Dialogs (December 3, 2021)
    Submit your review here
12. Continual Learning in Task-Oriented Dialogue Systems (December 10, 2021)
    Submit your review here
13. Single-dataset Experts for Multi-dataset Question Answering (December 17, 2021)
    Submit your review here