Hallucination-resistant multimodal content generation through knowledge graph-based reinforcement learning

doi:10.1016/j.inffus.2025.103783

Hallucination-resistant multimodal content generation through knowledge graph-based reinforcement learning

2025-09-30

0

PRE

AI

DOI：10.1016/j.inffus.2025.103783

Original

Original request for help

Save

Abstract

En 中文

• The prompt generation method based on the knowledge graph realizes the precise knowledge constraint of the multimodal large model. • The reasoning optimization method based on reinforcement learning has improved the model performance. • Chain of thought provides interpretable reasoning paths.

Hallucination-resistant multimodal content generation through knowledge graph-based reinforcement learning

Abstract

Journal

Organization