Post-Training Denoising of User Profiles with LLMs in Collaborative Filtering Recommendation

2026-01-25

AI Summary

This work addresses the detrimental impact of noise in implicit feedback on collaborative filtering performance. Existing denoising approaches often incur high costs by requiring model modifications, retraining, or additional data. To overcome these limitations, the authors propose a post-training denoising framework that leverages large language models (LLMs) to cleanse user interaction histories for the first time. Through carefully designed prompt engineering, the method inputs user profiles, candidate items, and initial rankings to guide the LLM in identifying and removing noisy interactions—without altering the underlying recommendation model or necessitating retraining. Extensive experiments across three benchmark datasets, integrating four open- and closed-source LLMs, demonstrate performance improvements of up to 13%, confirming the framework’s effectiveness and generalizability.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Data Mining & Knowledge Management: Recommender Systems
• 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: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Implicit feedback -- the main data source for training Recommender Systems (RSs) -- is inherently noisy and has been shown to negatively affect recommendation effectiveness. Denoising has been proposed as a method for removing noisy implicit feedback and improving recommendations. Prior work has focused on in-training denoising, however this requires additional data, changes to the model architecture and training procedure or fine-tuning, all of which can be costly and data hungry. In this work, we focus on post-training denoising. Different from in-training denoising, post-training denoising does not involve changing the architecture of the model nor its training procedure, and does not require additional data. Specifically, we present a method for post-training denoising user profiles using Large Language Models (LLMs) for Collaborative Filtering (CF) recommendations. Our approach prompts LLMs with (i) a user profile (user interactions), (ii) a candidate item, and (iii) its rank as given by the CF recommender, and asks the LLM to remove items from the user profile to improve the rank of the candidate item. Experiments with a state-of-the-art CF recommender and 4 open and closed source LLMs in 3 datasets show that our denoising yields improvements up to 13% in effectiveness over the original user profiles. Our code is available at https://github.com/edervishaj/denoising-user-profiles-LLM.

Problem

• implicit feedback
• denoising
• collaborative filtering
• post-training
• user profiles

Innovation

• post-training denoising
• Large Language Models
• Collaborative Filtering
• implicit feedback
• user profile