New DESRec framework personalizes latent reasoning for sequential recommendations

Researchers have developed DESRec, a new framework for sequential recommendation systems that aims to improve user modeling by personalizing the latent reasoning process. Unlike existing methods that use a fixed number of reasoning steps, DESRec dynamically selects the optimal reasoning depth for each user's history. This is achieved through a reasoning step selection module and a progressive preference extraction mechanism that segments user history into periods, summarizing each into a latent milestone. Experiments on six public datasets and four backbone recommenders showed DESRec outperforming eight baseline models. AI

IMPACT Could improve the accuracy and personalization of recommendation engines across various platforms.

RANK_REASON Research paper detailing a new framework for sequential recommendation systems. [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
• DESRec
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →