ML Anthology
Authors
Search
About
Pandey, Punya Syon
1 publications
ICLR
2026
SocialHarmBench: Revealing LLM Vulnerabilities to Socially Harmful Requests
Punya Syon Pandey
,
Lê Hải Sơn
,
Devansh Bhardwaj
,
Zhijing Jin