arXiv:2607. 00402v1 Announce Type: cross Abstract: Safety alignment of text-to-image (T2I) diffusion models aims to suppress harmful generations while preserving utility on benign prompts.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.
arXiv:2503. 11832v5 Announce Type: replace Abstract: Recent vision language models (VLMs) have made remarkable strides in generative modeling with multimodal inputs, particularly text and images.
Text-to-image diffusion models have achieved high visual fidelity and broad adoption, but remain vulnerable to safety violations when adversaries exploit them to synthesize illicit content. Existing alignment paradigms, from input sanitization to structural feature pruning, are largely organized around unsafe concepts explicitly exposed during filtering, editing, or localization.
arXiv:2605. 08245v4 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) increasingly power high-stakes applications, from medical imaging to autonomous systems, yet they routinely hallucinate, confidently describing content not present in the input.
InGuard introduces an inner guardrail for text-to-image generation that operates within the model’s own representations, avoiding external classifiers. It grades prompts using the text encoder’s embeddings, modifies risky embeddings with SAGE to produce safe images, and employs a latent detector to halt generation early. Evaluated on the RevGen Safety Benchmark, InGuard achieves a 97.9–98.8% safety rate across five open-weight models while reducing benign disturbances, model parameters, and denoising steps.
CollageAttack is a black‑box jailbreak that exploits cross‑modal alignment flaws in text‑to‑image models by shifting harmful semantics into the image plane. It combines context‑relevant scenes, scene‑grounded textual carriers, and spatially distributed text fragments to produce images that reveal hidden harmful meaning. Experiments on both open‑weight and commercial models show success rates up to 86.0%, outperforming the strongest baseline by 18.5 percentage points and consistently generating more harmful outputs while preserving the source intent.
arXiv:2609.39688v1 Announce Type: cross Abstract: Multimodal encoders such as CLIP underlie many downstream systems, but their web-scale training data embed harmful associations that safety alignment...