The Temporal Moderation Gap: Text-to-Video Safety Filters Are Blind to Harm in Motion

arXiv:2609.26233 · cs.CV · Submitted 2026-08-19 · Read on arXiv

cs.CV

Submitted: 2026-08-19

Updated: 2026-08-19

Terminology

• AUC — 10× in this paper · explained in 2 episodes
• Benchmark — 7× in this paper · explained in 2 episodes
• Diffusion Model — 6× in this paper · explained in 8 episodes
• Generative Model — 6× in this paper · explained in 2 episodes
• Generative Models — 5× in this paper · explained in 2 episodes
• Computer Vision and Pattern Recognition — 4× in this paper
• Diffusion Models — 4× in this paper · explained in 6 episodes
• Prompt Engineering — 3× in this paper · explained in 4 episodes
• Fidelity — 2× in this paper · explained in 2 episodes
• Accountability — 1× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Alignment — 1× in this paper · explained in 1 episode
• CVF International Conference on Computer Vision — 1× in this paper
• Cross-Model Transfer — 1× in this paper · explained in 2 episodes
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• Qwen2.5-VL Technical Report
• Two Frames Matter: A Temporal Attack for Text-to-Video Model Jailbreaking
• LTX-Video: Realtime Video Latent Diffusion
• HunyuanVideo: A Systematic Framework For Large Video Generative Models
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• Wan: Open and Advanced Large-Scale Video Generative Models
• SPARK: Jailbreaking T2V Models by Synergistically Prompting Auditory and Recontextualized Knowledge
• InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
• TrajShield: Trajectory-Level Safety Mediation for Defending Text-to-Video Models Against Jailbreak Attacks

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models