[Adapt] [ADAPT][Seminar] ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning

[Yizhu Liu]

Hi Adapters,

In this seminar, I’ll give a talk about ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning, which is published on AAAI 2019.  		 	 	 		

In this paper, they present ATOMIC, an atlas of everyday commonsense reasoning, organized through 877k textual descriptions of inferential knowledge. Compared to existing resources that center around taxonomic knowledge, ATOMIC focuses on inferential knowledge organized as typed if-then relations with variables (e.g., “if X pays Y a compliment, then Y will likely return the compliment”). They propose nine if-then relation types to distinguish causes vs. effects, agents vs. themes, voluntary vs. involuntary events, and actions vs. mental states. By gen- eratively training on the rich inferential knowledge described in ATOMIC, the results show that neural models can acquire simple commonsense capabilities and reason about previously unseen events. 

Time: Wed 4:30pm

Venue: SEIEE 3-414

Best wishes,
Yizhu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20210224/e90f6374/attachment.html>