Increased LLM Vulnerabilities from Fine-tuning and Quantization.

Divyanshu Kumar Anurakt Kumar Sahil Agarwal Prashanth Harshangi

Published in: CoRR (2024)

Keyphrases

fine tuning
viable alternative
fine tune
information security
fine tuned
multiresolution
countermeasures
quantization error
color quantization
database
information retrieval
learning algorithm
general purpose
image coding
security risks
security vulnerabilities