Modern recommender systems increasingly leverage large language models (LLMs) for reranking to improve personalization. However, existing approaches face two key limitations: (1) heavy reliance on manually crafted prompts that are difficult to scale, and (2) inadequate handling of unstructured item metadata that complicates preference inference. We present AGP (Auto-Guided Prompt Refinement), a novel framework that automatically optimizes user profile generation prompts for personalized reranking. AGP introduces two key innovations: (1) position-aware feedback mechanisms for precise ranking correction, and (2) batched training with aggregated feedback to enhance generalization.
@article{arxiv.2504.03965,
title = {Automating Personalization: Prompt Optimization for Recommendation Reranking},
author = {Chen Wang and Mingdai Yang and Zhiwei Liu and Pan Li and Linsey Pang and Qingsong Wen and Philip Yu},
journal= {arXiv preprint arXiv:2504.03965},
year = {2025}
}