Master Thesis: Robustness of DeepStack Architecture on Vision–Language Models
Abstract Recent advances in multimodal Vision–Language Models (VLMs) have introduced new architectural mechanisms designed to strengthen fine-grained visual-textual alignment. Among these, the DeepStack architecture (see [1]), adopted for example in state-of-the-art models such as Qwen3-VL, represents a significant shift in how visual features are injected into the language model. Instead of feeding all visual tokens […]
Master Thesis: Multimodal Entropy Analysis and Manipulation for Vision–Language Models
Abstract In the context of Vision-Language Models (VLMs), adversarial examples (or attacks) are image and prompt pairs which, when fed to a target model, make it hallucinate, disclose information it shouldn’t, or misclassify information. In this framework, there has been a growing interest in entropy-based attacks (see e.g., [1, 2]); that is, attack strategies which […]