Researchers have introduced DIGCN, a novel Dual-granularity raw feature empowered Intermediate fusion Graph Convolutional Network designed to enhance multimodal recommendation systems. This new model addresses limitations in representation learning and modality fusion by leveraging Large Language Models (LLMs) to construct dual-granularity raw features directly. DIGCN also incorporates an intermediate fusion graph convolutional network with dynamic cross-modality weight correction to effectively capture and fuse cross-modality signals. Experiments on four real-world datasets demonstrate DIGCN's superior performance, particularly in scenarios with data sparsity and cold-start challenges, and highlight its extensibility. AI
IMPACT This research could lead to more accurate and personalized recommendation systems by better leveraging multimodal data.
RANK_REASON Academic paper detailing a new model and its performance. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- graph convolutional network
- Hugging Face
- Influence Flower
- large-language models
- Litmaps
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →