New framework NPRec combats popularity bias in LLM recommendations

Researchers have developed a new framework called NPRec to address the issue of popularity bias in large language model (LLM)-based recommendation systems. LLMs, trained on vast datasets, tend to favor popular items over genuine user preferences, a problem that is difficult to correct within the model's parameters. NPRec intervenes externally by using counterfactual reasoning to generate debiased textual guidelines that reflect actual user interests. These guidelines are then used at inference time to steer the LLM towards more accurate and personalized recommendations without altering the model's core parameters. Experiments on real-world datasets show NPRec improves recommendation accuracy, explanation quality, and its ability to reduce bias. AI

IMPACT This research could lead to more personalized and trustworthy recommendations by mitigating the inherent popularity bias in LLMs.

RANK_REASON The cluster is based on an academic paper detailing a new framework for LLM-based recommendation systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Guanrong Li
• Hugging Face
• Influence Flower
• Large language models
• NPRec
• ScienceCast

• paper
• model release
• product

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