A composite backdoor built from the difference of two specially trained task vectors stays effective under both task addition and subtraction in task arithmetic.
Bert: Pre-training of deep bidirectional transformers for language understanding
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors
A composite backdoor built from the difference of two specially trained task vectors stays effective under both task addition and subtraction in task arithmetic.