Co-AttenDWG: Coattentive Dimension-Wise Gating and Expert Fusion for Multimodal Offensive Content Detection
inforesearchPeer-Reviewed
research
Source: IEEE Xplore (Security & AI Journals)October 20, 2025
Summary
This paper presents Co-AttenDWG, a new method for detecting offensive content by combining text and images together. The approach uses coattention (a technique where two types of data pay attention to each other simultaneously), dimension-wise gating (a mechanism that selectively emphasizes important features at a detailed level), and expert fusion (combining predictions from multiple specialized models) to better understand how text and visual information relate to each other.
Classification
Attack SophisticationModerate
AI Component TargetedModel
Monthly digest — independent AI security research
Original source: http://ieeexplore.ieee.org/document/11207235
First tracked: May 1, 2026 at 02:03 PM
Classified by LLM (prompt v3) · confidence: 85%