PulseAugur / Brief
EN
LIVE 16:51:24

Brief

last 24h
[1/1] 222 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. Grounding-Driven Attack: Improving Encoder-based Adversarial Transferability against Large Vision-Language Models

    Researchers have developed a new method called Grounding-Driven Attack (GDA) to improve the transferability of adversarial attacks against large vision-language models (LVLMs). Existing attacks often assume similar encoder architectures, but GDA focuses on text-conditioned grounding regions, which are more stable across different LVLM architectures. The proposed method allocates perturbation budgets to these grounded regions and intensifies their disruption, demonstrating superior performance in black-box scenarios. AI

    IMPACT This research highlights a vulnerability in vision-language models and proposes a more effective attack strategy, potentially influencing future robustness evaluations and defense mechanisms.