AI Paper Atlas

Research map / Video / Video generation

RunawayEvil: Jailbreaking the Image-to-Video Generative Models

Songping Wang, Rufan Qian, Yueming Lyu, Qinglong Liu, Linzhuang Zou, Jie Qin, Songhua Liu, Caifeng Shan

CVPR 2026 · Video › Video generation · approach: Compositional text-to-video diffusion · 11 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
Use RunawayEvil instead of T2I jailbreak baselines (Sneaky, PGJ, DACA)I2V jailbreak · COCO2017, 4 I2V models x 3 judgesASR (%): average 87.6%, surpassing others by 58.5%-79%Improved
Add Strategy Memory Bank (ablation)RunawayEvil · COCO2017, Wan, Qwen-VL judgeASR (%): 66.0% -> 71.0%Improved
Add Strategy Customization Agent (RL) instead of random selection (ablation)RunawayEvil · COCO2017, Wan, Qwen-VL judgeASR (%): 75.0% -> 75.0%No clear effect
Add Strategy Exploration Agent (ablation)RunawayEvil · COCO2017, average over models and judgesASR (%): 74.2% -> 87.2%Improved
Attack both modalities jointly instead of text only (ablation)RunawayEvil · COCO2017, averageASR (%): 51.2% -> 87.2%Improved
Attack both modalities jointly instead of image only (ablation)RunawayEvil · COCO2017, averageASR (%): 47.8% -> 87.2%Improved
Coordinate text and image attacks instead of attacking each in isolation (ablation)RunawayEvil · COCO2017, averageASR (%): 79.7% -> 87.2%Improved
Use adaptive steps with feedback instead of fixed 10 steps without feedback (ablation)RunawayEvil · COCO2017, Wan, Qwen-VL judgeASR (%): 68.0% -> 86.0%Improved
Use 10 steps instead of 1 step without feedback (ablation)RunawayEvil · COCO2017, Wan, Qwen-VL judgeASR (%): 48.0% -> 68.0%Improved
Add iteration penalty reward to success reward (ablation)SCA reward · COCO2017, Wan, Qwen-VL judgeASR (%): 80.0% -> 83.0%Improved
Add text stealth reward (ablation)SCA reward · COCO2017, Wan, Qwen-VL judgeASR (%): 83.0% -> 84.0%No clear effect
Add image stealth (LPIPS) reward (ablation)SCA reward · COCO2017, Wan, Qwen-VL judgeASR (%): 84.0% -> 86.0%No clear effect

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