Cite
Notes
Only stored in your browser.
Attribution
Towards Safety Reasoning in LLMs: AI-agentic Deliberation for Policy-embedded CoT Data Creation
arXiv 2025
Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
from 2 papers
Aram Galstyan
Charith Peris
Chongyu Fan
Jinghan Jia
Kai-Wei Chang
Mingyi Hong
Ninareh Mehrabi
Rahul Gupta
Richard Zemel
Sijia Liu