SHIFT learns and applies steering vectors to selected layers and timesteps in DiT models to suppress concepts, shift styles, or bias objects while keeping image quality and prompt adherence intact.
In: Findings of the Association for Computational Linguistics: ACL 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SHIFT: Steering Hidden Intermediates in Flow Transformers
SHIFT learns and applies steering vectors to selected layers and timesteps in DiT models to suppress concepts, shift styles, or bias objects while keeping image quality and prompt adherence intact.