Announcement
Two listed AI talks focus on manipulative LLM dialogue and simulated users
Maarten Sap listed presentations by Jocelyn Shen and Xuhui Zhou for Session 1 at 11 a.m. PDT.
TLDR
In October 6 posts, Maarten Sap listed two Session 1 presentations for 11 a.m. PDT. Jocelyn Shen’s talk asks whether it’s possible to predict when manipulative LLM dialogue will change a person’s beliefs. Xuhui Zhou’s asks how well simulated users stand in for real people when evaluating agents.
Combined views
900
1 Source, first seen ago