Xiao Zhang
Xiao Zhang
About
Research
Publication
Student
Teaching
Service
Contact
Light
Dark
Automatic
Subpopulation Data Poisoning
Oct 18, 2025
GREAT: Generalizable Backdoor Attacks in RLHF via Emotion-Aware Trigger Synthesis
we develop a novel framework for crafting generalizable backdoors in RLHF through emotion-aware trigger synthesis
Subrat Kishore Dutta
,
Yuelin Xu
,
Piyush Pant
,
Xiao Zhang
PDF
Cite
ArXiv
Cite
×