Metraipolis › archive › dc_e93121dc7501
TempJail: Temporal Jailbreak Attacks against Image-to-Video Generation Models — arXiv:2608.26971
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.26971
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T18:54:02.508Z
- SHA-256 of the archived text
155ed40a548c1c1848469c18a09b586211126d8866c095b512455a2de07a08c8- Stable URL of this page
https://metraipolis.com/documento/dc_e93121dc7501
How to cite
TempJail: Temporal Jailbreak Attacks against Image-to-Video Generation Models — arXiv:2608.26971 — arXiv (en), https://arxiv.org/abs/2608.26971 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e93121dc7501 (Metraipolis archive dc_e93121dc7501, sha256 155ed40a548c1c1848469c18a09b586211126d8866c095b512455a2de07a08c8)
BibTeX
@misc{dce93121dc7501,
title = {TempJail: Temporal Jailbreak Attacks against Image-to-Video Generation Models — arXiv:2608.26971},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.26971},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e93121dc7501; sha256 155ed40a548c1c1848469c18a09b586211126d8866c095b512455a2de07a08c8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26971. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In recent years, image-to-video (I2V) generation models have made remarkable progress in subject consistency and temporal coherence, enabling high quality video synthesis. However, these advances also introduce new safety risks. Existing studies mainly focus on jailbreak attacks involving single frame violations, while largely overlooking the temporal dimension unique to video generation models. In this paper, we investigate three attack scenarios and uncover a temporal vulnerability in I2V systems: unsafe semantics may emerge not from a single frame, but from semantic composition over time. We further identify two key challenges in such attacks: temporal abstraction and semantic camouflage. To address these issues, we propose TempJail, a novel temporal jailbreak framework for I2V systems. For temporal abstraction, we decompose a target malicious caption into an initial frame visual condition and a temporal text instruction. For semantic camouflage, on the image side we model semantic injection as controlled latent perturbation in diffusion sampling and introduce gradient guidance from pretrained encoders. On the text side, we rewrite the caption into an innocuous ``subject-action-scene'' template that bypasses safety filters while preserving temporal guidance. In the black-box inference phase, these two modalities jointly enable malicious semantics to be gradually triggered over time. Experiments on closed-source commercial models, including Kling, Seedance, Veo and PixVerse, show that TempJail improves attack success rate over prior state-of-the-art methods by 23.3\% under GPT-5.2 evaluation and 22.0\% under human evaluation. Our codes are available at \href{ this https URL }{GitHub}.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.MM).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2608.26971 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T18:31:14.486Z dalla pagina https://arxiv.org/catchup/cs.MM/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).