We investigate various prompting strategies for enhancing personalizedcontent recommendation performance with large language models (LLMs) throughinput augmentation. Our proposed approach, termed LLM-Rec, encompasses fourdistinct prompting strategies: (1) basic prompting, (2) recommendation-drivenprompting, (3) engagement-guided prompting, and (4) recommendation-driven +engagement-guided prompting. Our empirical experiments show that combining theoriginal content description with the augmented input text generated by LLMusing these prompting strategies leads to improved recommendation performance.This finding highlights the importance of incorporating diverse prompts andinput augmentation techniques to enhance the recommendation capabilities withlarge language models for personalized content recommendation.
URL
Affiliations
Abstract
Translation (by gpt-3.5-turbo)
Summary (by gpt-3.5-turbo)