LLM4DSR: Leveraing Large Language Model for Denoising Sequential Recommendation

2024-08-15

arXiv.org

AI Summary

To address performance degradation in sequential recommendation caused by noisy user interactions in behavioral sequences, this paper proposes an unsupervised, large language model (LLM)-driven sequence denoising framework. Methodologically, it introduces (1) a novel self-supervised fine-tuning task to activate the LLM’s semantic capability for implicit noise identification; (2) an uncertainty estimation module to effectively mitigate LLM hallucination; and (3) a model-agnostic denoising result reuse mechanism compatible with arbitrary downstream sequential recommenders. Extensive experiments across multiple benchmark datasets demonstrate consistent improvements: mainstream sequential recommendation models achieve average gains of 3.2%–7.8% in Recall@10 and NDCG@10. The framework establishes a new paradigm for noise-robust sequential recommendation, advancing both interpretability and reliability in real-world deployment scenarios.

Technology Category

• Reasoning under Uncertainty: Sequential Decision Making
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Sequential Recommenders generate recommendations based on users' historical interaction sequences. However, in practice, these collected sequences are often contaminated by noisy interactions, which significantly impairs recommendation performance. Accurately identifying such noisy interactions without additional information is particularly challenging due to the absence of explicit supervisory signals indicating noise. Large Language Models (LLMs), equipped with extensive open knowledge and semantic reasoning abilities, offer a promising avenue to bridge this information gap. However, employing LLMs for denoising in sequential recommendation presents notable challenges: 1) Direct application of pretrained LLMs may not be competent for the denoising task, frequently generating nonsensical responses; 2) Even after fine-tuning, the reliability of LLM outputs remains questionable, especially given the complexity of the denoising task and the inherent hallucinatory issue of LLMs. To tackle these challenges, we propose LLM4DSR, a tailored approach for denoising sequential recommendation using LLMs. We constructed a self-supervised fine-tuning task to activate LLMs' capabilities to identify noisy items and suggest replacements. Furthermore, we developed an uncertainty estimation module that ensures only high-confidence responses are utilized for sequence corrections. Remarkably, LLM4DSR is model-agnostic, allowing corrected sequences to be flexibly applied across various recommendation models. Extensive experiments validate the superiority of LLM4DSR over existing methods.

Problem

• Identifying noisy interactions in sequential recommendations without explicit signals
• Overcoming LLMs' limitations in generating reliable denoising outputs
• Enhancing recommendation performance by correcting contaminated user interaction sequences

Innovation

• Self-supervised fine-tuning for LLM denoising
• Uncertainty estimation for reliable corrections
• Model-agnostic sequence correction approach