FlowGuard detects unsafe content during diffusion image generation via linear latent decoding and curriculum learning, outperforming prior methods by over 30% F1 while reducing GPU memory by 97% and projection time to 0.2 seconds.
Prompting4debugging: Red-teaming text-to-image diffusion models by finding problematic prompts
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 6roles
background 1polarities
background 1representative citing papers
TICoE achieves more precise and faithful concept erasure in text-to-image models by collaborating text and image data through a convex manifold and hierarchical learning, outperforming prior methods.
A training-free double-projection linear transformation erases target concepts from generative models by computing a proxy projection then applying a constrained update in the left null space of known directions.
EGLOCE erases target concepts in diffusion models at inference time by optimizing latents with dual energy guidance that repels unwanted concepts while retaining prompt alignment.
SPACE induces sparsity in cross-attention parameters via closed-form iterative updates to erase target concepts more effectively than dense baselines in large diffusion models.
citing papers explorer
-
FlowGuard: Towards Lightweight In-Generation Safety Detection for Diffusion Models via Linear Latent Decoding
FlowGuard detects unsafe content during diffusion image generation via linear latent decoding and curriculum learning, outperforming prior methods by over 30% F1 while reducing GPU memory by 97% and projection time to 0.2 seconds.
-
Beyond Text Prompts: Precise Concept Erasure through Text-Image Collaboration
TICoE achieves more precise and faithful concept erasure in text-to-image models by collaborating text and image data through a convex manifold and hierarchical learning, outperforming prior methods.
-
Closed-Form Concept Erasure via Double Projections
A training-free double-projection linear transformation erases target concepts from generative models by computing a proxy projection then applying a constrained update in the left null space of known directions.
-
EGLOCE: Training-Free Energy-Guided Latent Optimization for Concept Erasure
EGLOCE erases target concepts in diffusion models at inference time by optimizing latents with dual energy guidance that repels unwanted concepts while retaining prompt alignment.
-
Empty SPACE: Cross-Attention Sparsity for Concept Erasure in Diffusion Models
SPACE induces sparsity in cross-attention parameters via closed-form iterative updates to erase target concepts more effectively than dense baselines in large diffusion models.
- FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models