Selective LLM-Guided Regularization for Enhancing Recommendation Models

TL;DR

This paper proposes a selective regularization framework that leverages large language models for recommendation tasks, activating LLM guidance only when it is deemed reliable, thereby improving accuracy especially in cold start and long tail scenarios.

Contribution

It introduces a novel, model-agnostic, and efficient selective LLM-guided regularization method that activates guidance based on a trainable gating mechanism, overcoming limitations of previous global distillation approaches.

Findings

Improves recommendation accuracy across multiple datasets.

Enhances performance in cold start and long tail regimes.

Outperforms global distillation baselines.

Abstract

Large language models provide rich semantic priors and strong reasoning capabilities, making them promising auxiliary signals for recommendation. However, prevailing approaches either deploy LLMs as standalone recommender or apply global knowledge distillation, both of which suffer from inherent drawbacks. Standalone LLM recommender are costly, biased, and unreliable across large regions of the user item space, while global distillation forces the downstream model to imitate LLM predictions even when such guidance is inaccurate. Meanwhile, recent studies show that LLMs excel particularly in re-ranking and challenging scenarios, rather than uniformly across all contexts.We introduce Selective LLM Guided Regularization, a model-agnostic and computation efficient framework that activates LLM based pairwise ranking supervision only when a trainable gating mechanism informing by user history length, item popularity, and model uncertainty predicts the LLM to be reliable. All LLM scoring is performed offline, transferring knowledge without increasing inference cost. Experiments across multiple datasets show that this selective strategy consistently improves overall accuracy and yields substantial gains in cold start and long tail regimes, outperforming global distillation baselines.