[Adapt] [Seminar]Ethical Issues in Dialogue Systems

[贾琪]

Hi Adapters,

Data-driven models inevitably learned some undesired features from large-scale datasets, such as offensive languages, gender discrimination, etc. Such ethical issues are gaining more and more attention in recent years, especially in dialogue systems. In this talk, I'll introduce some representative unsafe problems and a social bias detection framework.

It's also an interesting topic, hope you enjoy it.

Time: Wed 4:00pm
Venue: SEIEE 3-414
Best Wish,
Angel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20220302/5e24013d/attachment.htm>