2 papers
cs.CV2026
FreqDoor: A Hidden Trojan in the Frequency Domain for Backdoor Attacks on Vision-Language Models
Yasir Arafat Prodhan, Sadad Hasan, Mohammed Imamul Hassan Bhuiyan
Vision-language models (VLMs) have recently shown excellent progress in open-ended image-to-text generation. However, their multimodal nature makes them persistently vulnerable to…
eess.IV2025
SF-Net: A Multi-Modal Approach to Medical Image Classification via Spatial-Spectral Summarizer Fusion Network
Md. Saiful Bari Siddiqui, Mohammed Imamul Hassan Bhuiyan
Convolutional Neural Networks have become a cornerstone of medical image analysis due to their proficiency in learning hierarchical spatial features. However, this focus on a singl…