Hybrid explicit and implicit encoding for multi-view representation learning

doi:10.1016/j.patcog.2026.113584

Hybrid explicit and implicit encoding for multi-view representation learning

2026-03-25

0

PRE

AI

DOI：10.1016/j.patcog.2026.113584

Original

Original request for help

Save

Abstract

En 中文

• Propose HEIM: a novel hybrid encoder combining implicit and explicit encoding. • Introduce an adaptive information flow for scalable and interpretable encoding. • Develop a mutual information fusion strategy with cross-view regularization. • Design a dual-level alignment module for compactness and discrimination.

Keywords:

Hybrid encoding

Multi-view representation learning

Adaptive information flow

Mutual information fusion

Dual-level alignment

Hybrid explicit and implicit encoding for multi-view representation learning

Abstract

Journal

Organization

Cited Papers