REGEN: A Dataset and Benchmarks with Natural Language Critiques and Narratives
Topics: AI Mode, AIOverviews, LLM Readability, LLMO / GEO, Personalization, Shopping
The Google research paper introduces REGEN (Reviews Enhanced with GEnerative Narratives), a dataset designed to improve conversational recommendation systems by incorporating user critiques and rich narratives. It aims to enhance the capabilities of Large Language Models (LLMs) in generating informed recommendations based on users’ prior interactions. The paper presents a joint modeling framework, LUMEN, which integrates critiques, recommendations, and narratives, showcasing improved performance in generating personalized recommendations that resonate with user preferences.