Multi-Level Collaborative Optimization for Multi-Modal Media Manipulation Detection and Grounding

Wenwen Su

doi:10.1016/j.inffus.2026.104773

Multi-Level Collaborative Optimization for Multi-Modal Media Manipulation Detection and Grounding

2026-09-12

0

PRE

AI

DOI：10.1016/j.inffus.2026.104773

Original

Original request for help

Save

Abstract

En 中文

• A new task-collaboration view is proposed for multimodal media forensics. • A three-level collaboration path links features, semantics, and decisions. • Shared semantic weighting guides detection and grounding to consistent evidence. • Decision-aware refinement suppresses noisy visual grounding responses.

Keywords:

Multi-modal learning

Media manipulation

Deepfake detection

Task collaboration

Semantic consistency

Multi-Level Collaborative Optimization for Multi-Modal Media Manipulation Detection and Grounding

Abstract

Journal

Organization

Cited Papers