Contrastive Inversion disentangles the common concept from per-image auxiliary tokens via InfoNCE loss, then fine-tunes only the target cross-attention pathway, matching DisenBooth's numbers while claiming better qualitative editability.
Classifier-free diffusion guidance
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
method 1polarities
use method 1representative citing papers
citing papers explorer
-
Comparison Reveals Commonality: Customized Image Generation through Contrastive Inversion
Contrastive Inversion disentangles the common concept from per-image auxiliary tokens via InfoNCE loss, then fine-tunes only the target cross-attention pathway, matching DisenBooth's numbers while claiming better qualitative editability.