From Entity Reliability to Clean Feedback: An Entity-Aware Denoising Framework Beyond Interaction-Level Signals

Implicit feedback is central to modern recommender systems but is inherently noisy, often impairing model training and degrading user experience. At scale, such noise can mislead learning processes, reducing both recommendation accuracy and platform value. Existing denoising strategies typically overlook the entity-specific nature of noise while introducing high computational costs and complex hyperparameter tuning. To address these challenges, we propose \textbf{EARD} (\textbf{E}ntity-\textbf{A}ware \textbf{R}eliability-\textbf{D}riven Denoising), a lightweight framework that shifts the focus from interaction-level signals to entity-level reliability. Motivated by the empirical observation that training loss correlates with noise, EARD quantifies user and item reliability via their average training losses as a proxy for reputation, and integrates these entity-level factors with interaction-level confidence. The framework is \textbf{model-agnostic}, \textbf{computationally efficient}, and requires \textbf{only two intuitive hyperparameters}. Extensive experiments across multiple datasets and backbone models demonstrate that EARD yields substantial improvements over state-of-the-art baselines (e.g., up to 27.01\% gain in NDCG@50), while incurring negligible additional computational cost. Comprehensive ablation studies and mechanism analyses further confirm EARD's robustness to hyperparameter choices and its practical scalability. These results highlight the importance of entity-aware reliability modeling for denoising implicit feedback and pave the way for more robust recommendation research.

• Publication year

  2025

• Venue

  Unknown venue

• Publication date

  2025-08-14

• Fields of study

  Computer Science

• Identifiers
  arXiv 2508.10851
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Unleashing the Power of Large Language Model for Denoising Recommendation

  2025cited by this paper
• Personalized Denoising Implicit Feedback for Robust Recommender System

  2025cited by this paper
• Double Correction Framework for Denoising Recommendation

  2024cited by this paper
• Improving Multi-modal Recommender Systems by Denoising and Aligning Multi-modal Content and User Feedback

  2024cited by this paper
• SLED: Structure Learning based Denoising for Recommendation

  2023cited by this paper
• AutoDenoise: Automatic Data Instance Denoising for Recommendations

  2023cited by this paper
• Denoising Neural Network for News Recommendation with Positive and Negative Implicit Feedback

  2022cited by this paper
• Towards Robust Neural Graph Collaborative Filtering via Structure Denoising and Embedding Perturbation

  2022cited by this paper
• Learning to Denoise Unreliable Interactions for Graph Collaborative Filtering

  2022cited by this paper
• Self-Guided Learning to Denoise for Robust Recommendation

  2022cited by this paper
• Learning Robust Recommenders through Cross-Model Agreement

  2021influential reference
• Implicit Feedbacks are Not Always Favorable: Iterative Relabeled One-Class Collaborative Filtering against Noisy Interactions

  2021cited by this paper
• Denoising Implicit Feedback for Recommendation

  2020influential reference
• Are we really making much progress? A worrying analysis of recent neural recommendation approaches

  2019cited by this paper
• Co-teaching: Robust training of deep neural networks with extremely noisy labels

  2018cited by this paper
• Neural Collaborative Filtering

  2017cited by this paper
• Collaborative Denoising Auto-Encoders for Top-N Recommender Systems

  2016cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Personalized Ranking for Non-Uniformly Sampled Items

  2012cited by this paper
• Curriculum learning

  2009cited by this paper
• Collaborative Filtering for Implicit Feedback Datasets

  2008cited by this paper

• No citing papers are available for this paper.