Aggregated demonstrations improve LLM-based sequential recommendation by combining multiple users into one demonstration. The study explores factors like instruction format, task consistency, and demonstration selection. LLMSRec-Syn outperforms existing methods on three datasets. The number of member users in aggregated demonstrations impacts performance. LLMSRec-Syn competes with supervised methods and benefits from powerful LLMs.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Lei Wang,Ee-... at arxiv.org 03-18-2024
https://arxiv.org/pdf/2403.10135.pdfDeeper Inquiries