- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 1
Read review
Paper summary
This paper studies visual prompt attacks against I2V generation models. Visual cues such as arrows, sketches, poses, and frame transitions can induce unsafe video content. They introduce VVA-Bench, containing 452 examples on six formats for attacks. Authors propose VPA-Guard, a retrieval-augmented defense that uses attack patterns and self-evolving defense framework. VPA-Guard acheieves 44.2% ASR drop on evaluation in four I2V models.
Critical strengths
-
The paper considers an important safety problem in I2V generation. The benchmark covers several visual control mechanisms, including temporal transitions, which are not adequately captured by text-only safety evaluations.
-
The evaluation covers multiple commercial models and includes comparisons on text v.s. visual, controls on benign samples, and human calibration of the LLM-as-a-judge.
-
The paper is well-written and easy-to-follow.
Critical weaknesses
-
I have a serious concern about the ASR calculation. In Figure 4, the reported ASRs appear to be exactly , where means refusal. However, Table 10 explicitly defines as Safe and as Borderline Safe. For example, Wan 2.7 reports 100% ASR, where Figure 4 shows that more than 50% of its outputs fall into these two safe categories. This suggests that the paper may be counting any non-refusal as a successful attack, without checking whether the generated video is actually harmful. This directly contradicts the stated ASR definition and calls the main experimental conclusions into question. I also find the custom harmfulness score in Appendix F.2 difficult to justify, since even safe outputs can receive substantial positive scores.
-
The novelty and defense evaluation are marginal. Visual instruction attacks on I2V models have already been studied in recent work like Visual Instruction Injection [a]. The proposed defense mainly combines standard embedding retrieval, few-shot safety prompting, and database updates. Furthermore, there is no convincing comparison against simple safety prompting or direct VLM-based safety screening. The current ablation does not isolate the contribution of retrieval or rationale prompting.
Ref: [a] Zheng et al. VII: Visual Instruction Injection for Jailbreaking Image-to-Video Generation Models.
Questions for authors
-
How exactly is ASR computed? Why does it appear to equal the non-refusal rate in Figure 4? Could the authors recompute the main results using actual harmful outputs, for example ?
-
How much improvement remains when comparing VPA-Guard against a simple safety prompt or a direct VLM safety classifier under matched benign rejection rates?