analiza aiqbench
This research addresses a critical flaw in LLM-based recommender systems where models default to suggesting popular items rather than genuine user preferences. The proposed NPRec framework mitigates this popularity bias by injecting counterfactual reasoning guidelines at inference time, effectively separating intrinsic interests from trend-driven conformity without requiring model retraining. This approach enhances both recommendation accuracy and the faithfulness of generated explanations, marking a significant step toward more trustworthy generative AI applications.
Neutralizing Popularity Bias in LLM-based Recommendation via Counterfactual Reasoning Guidelines
arXiv:2503.08051v2 Announce Type: replace Abstract: In the era of generative AI, recommender systems are moving from precise prediction to trustworthy generation. Large language models (LLMs) support this shift by inferring user interests and producing natural-language explanations. However, LLM-based recommendation suffers from a fundamental obstacle: popularity bias. Through pre-training on massive corpora, LLMs tend to rely on global statistics and trend signals, yielding recommendations that follow popularity rather than genuine preference. As this bias is entangled in model parameters and is hard to remove directly, we propose Neutralizing Popularity Bias in LLM-based Recommendation via Counterfactual Reasoning Guidelines (NPRec), a model-agnostic framework that mitigates popularity bias through external semantic intervention. NPRec performs counterfactual refinement to causally separate intrinsic user interests from popularity-driven conformity, producing debiased textual guidelines that reflect actual user preferences. These guidelines are injected at inference time to shift the LLM from unconstrained generation to guided reasoning, without any parameter updates. Serving as explicit premises, they both ground faithful explanations and improve recommendation quality. Extensive experiments on three real-world datasets demonstrate that NPRec achieves promising performance in recommendation accuracy, explanation quality, and debiasing capability.
Abstract:In the era of generative AI, recommender systems are moving from precise prediction to trustworthy generation. Large language models (LLMs) support this shift by inferring user interests and producing natural-language explanations. However, LLM-based recommendation suffers from a fundamental obstacle: popularity bias. Through pre-training on massive corpora, LLMs tend to rely on global statistics and trend signals, yielding recommendations that follow popularity rather than genuine preference. As this bias is entangled in model parameters and is hard to remove directly, we propose Neutralizing Popularity Bias in LLM-based Recommendation via Counterfactual Reasoning Guidelines (NPRec), a model-agnostic framework that mitigates popularity bias through external semantic intervention. NPRec performs counterfactual refinement to causally separate intrinsic user interests from popularity-driven conformity, producing debiased textual guidelines that reflect actual user preferences. These guidelines are injected at inference time to shift the LLM from unconstrained generation to guided reasoning, without any parameter updates. Serving as explicit premises, they both ground faithful explanations and improve recommendation quality. Extensive experiments on three real-world datasets demonstrate that NPRec achieves promising performance in recommendation accuracy, explanation quality, and debiasing capability.
Artificial Intelligence (cs.AI)
arXiv:2503.08051 [cs.AI]
(or arXiv:2503.08051v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2503.08051 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 271 cuvinte
Subiecte conexe
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · 13 oct. 2026
Rational Agentic Maximalist Philosophies
LessWrong · 17 iun. 2026
Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs
LessWrong · 27 iul. 2026
A Red Line and Oversight Framework for Government AI Contracts
AI Alignment Forum · 18 iul. 2026