More Web Proxy on the site http://driver.im/

default search action

combined dblp search
author search
venue search
publication search

ask others

"Soft Prompt Threats: Attacking Safety Alignment and Unlearning in ..."

Leo Schwinn et al. (2024)

> Home

Details and statistics

DOI: 10.48550/ARXIV.2402.09063

access: open

type: Informal or Other Publication

metadata version: 2024-02-19

- view
  - electronic edition via DOI (open access)
  authority control:
- export record
  dblp key:
  - journals/corr/abs-2402-09063
- ask others
- share record
  persistent URL:
  - https://dblp.org/rec/journals/corr/abs-2402-09063
Leo Schwinn, David Dobre, Sophie Xhonneux, Gauthier Gidel, Stephan Günnemann:
Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space. CoRR abs/2402.09063 (2024)

manage site settings

To protect your privacy, all features that rely on external API calls from your browser are turned off by default. You need to opt-in for them to become active. All settings here will be stored as cookies with your web browser. For more information see our F.A.Q.