Can I read LLM Vulnerability: Small Sample Poisoning on EtoBox?
LLM Vulnerability: Small Sample Poisoning by Marcos Merino is a document available to read on EtoBox.
What is LLM Vulnerability: Small Sample Poisoning about?
A study by Anthropic, the UK AI Security Institute, and the Alan Turing Institute reveals that as few as 250 malicious documents can successfully backdoor large language models (LLMs) of any size, challenging the assumption that attackers need a percentage of training data. The research indicates that the effectiveness of poisoning attacks depends on the absolute number of poisoned documents rather than their proportion to the total training data. This finding highlights the need for further investigation i
- Author
- Marcos Merino
- Language
- EN