Towards a Human-like Open-Domain Chatbot

dc.contributor.authorAdiwardana, Daniel
dc.contributor.authorLuong, Minh-Thang
dc.contributor.authorSo, David R.
dc.contributor.authorHall, Jamie
dc.contributor.authorFiedel, Noah
dc.contributor.authorThoppilan, Romal
dc.contributor.authorYang, Zi
dc.contributor.authorKulshreshtha, Apoorv
dc.contributor.authorNemade, Gaurav
dc.contributor.authorLu, Yifeng
dc.contributor.authorLe, Quoc V.
dc.date.accessioned2025-06-02T13:34:32Z
dc.date.available2025-06-02T13:34:32Z
dc.date.issued2020-01-27
dc.description2.6B 파라미터 Meena 모델을 제안하며, SSA(Sensibleness and Specificity Average)라는 새로운 평가 지표를 통해 다중턴 대화 성능을 평가합니다. 최고 성능 모델은 SSA 79%를 기록하며, 인간 수준(86%)에 근접하는 가능성을 제시합니다 ©2020 Google Research
dc.description.abstractWe present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6B parameter neural network is simply trained to minimize perplexity of the next token. We also propose a human evaluation metric called Sensibleness and Specificity Average (SSA), which captures key elements of a human-like multi-turn conversation. Our experiments show strong correlation between perplexity and SSA. The fact that the best perplexity end-to-end trained Meena scores high on SSA (72% on multi-turn evaluation) suggests that a human-level SSA of 86% is potentially within reach if we can better optimize perplexity. Additionally, the full version of Meena (with a filtering mechanism and tuned decoding) scores 79% SSA, 23% higher in absolute SSA than the existing chatbots we evaluated.
dc.description.sponsorshipGoogle Research (Brain Team)
dc.identifier.urihttps://arxiv.org/abs/2001.09977
dc.identifier.urihttp://data.inu.ac.kr/handle/123456789/1961
dc.language.isoen_US
dc.publisherarXiv
dc.subjectMeena
dc.subjectOpen-Domain Chatbot
dc.subjectSSA
dc.subjectMulti-turn Conversation
dc.subjectDialogue Systems
dc.titleTowards a Human-like Open-Domain Chatbot
dc.typeArticle

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
2001.09977v3.pdf
Size:
1011.25 KB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
97 B
Format:
Item-specific license agreed to upon submission
Description: