arXiv:2607. 24101v1 Announce Type: cross Abstract: Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models.
Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models. Existing evaluations, however, mostly study targets that dominate the whole image, such as styles, broad object categories, or portrait-like identities, leaving company logos comparatively underexamined.
arXiv:2609.37127v1 Announce Type: new Abstract: Establishing unbranding as a critical practice to prevent visual logos from acquiring negative connotations is standard in image generation. Large Lang...
arXiv:2609.01433v1 Announce Type: new Abstract: Concept erasure aims to suppress unsafe, privacy-sensitive, or undesirable generations in text-to-image diffusion models while preserving benign semant...
arXiv:2610.20603v1 Announce Type: new Abstract: We present CommerceGen-Studio, an end-to-end framework for generating commercial-grade main images for e-commerce Stock Keeping Units (SKUs). Our frame...
arXiv:2607. 23492v1 Announce Type: cross Abstract: Concept erasure techniques (CETs) edit text-to-image diffusion models to erase undesired targets such as NSFW content or copyrighted styles, while preserving model utility on benign concepts.
arXiv:2412.07658v3 Announce Type: replace-cross Abstract: Text-to-image diffusion models have rapidly transitioned from research prototypes to widely deployed tools. Along with their impressive capab...
We present CommerceGen-Studio, an end-to-end framework for generating commercial-grade main images for e-commerce Stock Keeping Units (SKUs). Our framework is organized along a three-stage pipeline of...
GRACE is a new framework for concept erasure in text-to-image diffusion models that uses a semantically weighted sensitive subspace to guide localized interventions and lightweight subspace-constrained adapters to avoid global semantic disruption. It replaces manual counterfactual prompts with an automatically decoupled safe-anchor mechanism and controls intervention strength through an energy-driven dynamic gating system. Experiments show GRACE improves NSFW reduction by 17.86% over five state-of-the-art methods while also reducing target CLIP Score and FID, indicating stronger concept suppression with better preservation of generative quality.
RASteer is a training‑free method for concept erasure in text‑to‑image diffusion models that retains other concepts. It constructs a retain subspace from concepts to preserve, then removes components aligned with this subspace from the erasure direction using Retain‑Orthogonal Steering (ROS). Overlap‑Adaptive Calibration (OAC) further adjusts the removal of shared components at each layer and denoising step, balancing target erasure with concept preservation. Experiments show RASteer matches or outperforms existing activation steering and weight editing baselines on unsafe‑content, instance, and artistic‑style erasure tasks across multiple backbones and benchmarks.
arXiv:2509. 22415v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have achieved strong vision-language performance, yet their token-level visual evidence remains difficult to inspect.
The paper introduces Dynamic Semantic Steering (DSS), a training‑free, inference‑time defense for robust concept erasure in text‑to‑image diffusion models. DSS models local semantic neighborhoods geometrically, automatically identifies benign semantic anchors, and applies context‑aware, constrained feature correction using cross‑attention signals. Experiments show DSS achieves an average erasure rate of 91.0%, outperforming prior defenses while reducing semantic drift and preserving generation fidelity.
The paper introduces DOGS, a method that reframes logo generation as sampling within a structured design space rather than text rewriting. It constructs a typed, graph-structured grammar from real-world logos and uses a GFlowNet sampler to produce design graphs weighted by recognizability, aesthetics, and originality, while filtering out potentially infringing or harmful elements. Experiments show that DOGS yields logos that are more recognizable, aesthetic, diverse, and less likely to infringe trademarks compared to nine baselines.