Safety Alignment in NLP Tasks: Weakly Aligned Summarization as an In-Context Attack.

Yu Fu Yufei Li Wen Xiao Cong Liu Yue Dong

Published in: ACL (1) (2024)

Keyphrases

artificial intelligence
natural language processing
context sensitive
context aware
information extraction
contextual information
countermeasures
learning algorithm
natural language
text mining
question answering
text analysis
soccer video
reference resolution