AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Latent Guard: a Safety Framework for Text-to-image Generation

Runtao Liu, Ashkan Khakzar, Jindong Gu, Qifeng Chen, Philip Torr, Fabio Pizzati

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Text generation and structured prediction · 46 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add safe prompts as negatives in the contrastive loss (ablation)Latent Guard · CoPro ID SynonymAUC: 0.607 -> 0.914Improved
use cross-attention instead of MLP in the Embedding Mapping Layer (ablation)Latent Guard · CoPro ID AdversarialAUC: 0.818 -> 0.908Improved
use cross-attention instead of MLP (ablation)Latent Guard · CoPro OOD ExplicitAUC: 0.947 -> 0.944Hurt
concept-matching setup instead of binary safe/unsafe classifier (ablation)Latent Guard · Unsafe Diffusionaccuracy: 0.737 -> 0.794Improved
use a larger blacklist at test time (10% to 100% of concepts) (ablation)Latent Guard · Unsafe Diffusionaccuracy: 0.548 -> 0.794Improved
increase attention heads from 1 to 16 (d=128) (ablation)Embedding Mapping Layer · CoPro fine-grained concept classificationaccuracy: 0.874 -> 0.931Improved
train Latent Guard embedding vs plain CLIP/BERT similarityconcept-prompt similarity · CoPro ID ExplicitAUC: 0.697 -> 0.985 (CLIPScore vs ours)Improved
Latent Guard prompt filtering vs LLM filtering before generationT2I pipeline · Unsafe DiffusionNudeNet+Q16 inappropriate rate: 0.138 -> 0.029Improved
Latent Guard vs LLM baseline on adversarial promptsCoPro Adversarial accuracy · CoPro ID Adversarialaccuracy: 0.867 vs 0.829Hurt

Later papers that compared with this method

1 later accepted paper reports Latent Guard as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy