[Dialog Response Selection] Do Response Selection Models Really Know What’s Next? Utterance Manipulation Strategies For Multi-turn Response Selection

NLP

코딩무민 2022. 5. 24. 17:18

1. 핵심 요약

user, system의 utterance history로 optimal response를 찾는 task
pre-trained language model이 다양한 NLP tasks에서 좋은 성능을 보이고 있음
→response selection tasks에서는 이를 dialog–response binary classification tasks로 품
위의 방법론은 sequential nature of multi-turn dialog system을 무시
위 논문
- the response selection task 하나는 불충분e.g. insertion, deletion, search 등 다양한 방법 제시
- → dialog coherence 유지에 도움이 됨
- : utterance manipulation strategies (UMS)
- UMS : self-supervised model → existing approaches에 쉽게 통합됨

Do Response Selection Models Really Know What’s Next? Utterance Manipulation Strategies For Multi-turn Response Selection

Abstract

coding-moomin.notion.site

[Dialog Response Selection] An Effective Domain Adaptive Post-Training Method for BERT in Response Selection (0)	2022.05.24
[SBERT 논문리뷰] Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks (0)	2022.05.24
[KBP] Zero-shot Slot Filling with DPR and RAG (0)	2022.05.24
[Numerical Reasoning] Have You Seen That Number? Investigating Extrapolation in Question Answering Models (0)	2022.05.24
[PromptBERT 논문리뷰] PromptBERT: Improving BERT Sentence Embeddings with Prompts (1)	2022.05.18

현재글[Dialog Response Selection] Do Response Selection Models Really Know What’s Next? Utterance Manipulation Strategies For Multi-turn Response Selection