This work introduces PaRMA, an evidence-based auditing framework that defines pathway-conditioned risk over two primary image-level interventions: pixel perturbation and rendered-text insertion and uncovers a see-but-not-deceived signature where native VLMs attend heavily to typographic overlays but visually reject the...
Jiaming Zhang, Zhe-Rui Li, Si-Qi Guo et al.· 0 citations
This work proposed PLUME, a training-free method that constructs a global update representation, activates memory evidence to form a local parameter view, and adaptively integrates their predictions during decoding, which demonstrated its effectiveness in sequential evolution settings.
GRAFT, an IPP-oriented post-training framework combining constraint-aware SFT and preference optimization is proposed, an IPP-oriented post-training framework combining constraint-aware SFT and preference optimization that raises the average IIF score and preserves general generation ability.
Zheng Nie, Zhe-Rui Li, Jia-Ming Zhang et al.· 0 citations
This survey examines the protective paradigm that has grown around this intervention point, and finds that most protections are still validated mainly against static or weakly adaptive adversaries, while evidence beyond controlled benchmarks remains scarce.
Jiaming Zhang, Bo-Yang Chen, Zhe-Rui Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.