arXiv:2607. 19061v1 Announce Type: cross Abstract: Hateful optical illusions expose a serious gap in current multimodal safety systems.
arXiv:2606. 09700v1 Announce Type: cross Abstract: Large language model (LLM)-powered content moderation systems have become a critical defense against harmful online content.
arXiv:2608. 05210v1 Announce Type: cross Abstract: Picture books and comics have long been used to disseminate hateful narratives because they are easily understood even by children, as exemplified by the notorious Nazi propaganda picture book \emph{Der Giftpilz}.
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.
The paper introduces MPS-Bench, a benchmark of 5,181 scenarios from 584 real-world images across 12 high-risk domains, each paired with a hidden user profile, to evaluate personalized safety in vision‑language models (VLMs). Eight leading VLMs were tested and found to almost always respond directly (86‑99%) without seeking missing context, scoring no higher than 2.6/5 on personalized safety. The authors identify a phenomenon called visual dominance, where visual information enters text representations early and suppresses textual risk signals, and propose PRISM, a lightweight input monitor that predicts when a query should be deferred, achieving 0.978 AUC and outperforming all tested models on the safety‑utility Pareto frontier.
The paper introduces TaintedPixels, a proactive video‑protection technique that embeds subtle, structured perturbations into the blue channel of facial regions. These perturbations remain invisible in the original video but become noticeable after black‑box manipulation, thereby flagging deepfakes. Experiments on three off‑the‑shelf manipulation tools and two detectors show that TaintedPixels achieves the highest forgery detection rate while keeping visual distortion minimal (LPIPS = 0.0042), and a human study confirms that protected videos are rarely suspected, whereas forgeries from protected sources are identified as fake far more often than those from unprotected sources.
arXiv:2511.18921v2 Announce Type: replace Abstract: Backdoor attacks undermine the reliability and trustworthiness of machine learning systems by injecting hidden behaviors that can be maliciously ac...
arXiv:2607. 22745v1 Announce Type: cross Abstract: Rapid advances in image generation are eroding the evidentiary value of visual content in settings where authenticity can affect public safety and personal reputation.
Large vision-language models (LVLMs) have recently shown immense potential in automated content moderation, sparking growing interest in developing harmful-video benchmarks. However, we identify two primary limitations in existing works: 1) The multi-layered characteristics of harmful videos are overlooked.
arXiv:2601. 11178v3 Announce Type: replace Abstract: Social media platforms are increasingly dominated by long-form multimodal content, where harmful narratives are constructed through a complex interplay of audio, visual, and textual cues.
The paper identifies a flaw in current text-to-video safety filters that rely on keyword prompts and per-frame checks, showing that these systems miss harms that arise solely from the ordering of frames. Experiments reveal that a significant portion of generated videos contain harmful content that is undetected because each individual frame appears safe, and prompt engineering does not mitigate this issue. The authors propose an order-aware detector that can distinguish harmful videos from benign shuffles, achieving an AUC of 0.74 compared to chance for per-frame checking.
The paper introduces ProKDA, a progressive knowledge-to-decision alignment framework for explainable hateful meme detection. ProKDA separates explanation generation and label prediction into three sequential training stages—background knowledge learning, hatefulness detection learning, and hatefulness boundary alignment—reducing task interference. Experiments on three public benchmarks demonstrate that ProKDA achieves state‑of‑the‑art detection performance while providing accurate, evidence‑supported explanations for moderation decisions.
arXiv:2610.18609v1 Announce Type: cross Abstract: Direct Preference Optimization (DPO) has become a widely adopted approach for aligning vision-language models (VLMs), yet the security of multimodal...