Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control

Luankang Zhang, Hao Wang, Zhongzhou Liu, Mingjia Yin, Yonghao Huang, Jiaqi Li, Wei Guo, Yong Liu, Huifeng Guo, Defu Lian, Enhong Chen

TL;DR

The Recursive Self-Improving Recommendation framework is proposed, a paradigm in which a model bootstraps its own performance without reliance on external data or teacher models, suggesting a scalable path forward for recommender systems and beyond.

Abstract

The scarcity of high-quality training data presents a fundamental bottleneck to scaling machine learning models. This challenge is particularly acute in recommendation systems, where extreme sparsity in user interactions leads to rugged optimization landscapes and poor generalization. We propose the Recursive Self-Improving Recommendation (RSIR) framework, a paradigm in which a model bootstraps its own performance without reliance on external data or teacher models. RSIR operates in a closed loop: the current model generates plausible user interaction sequences, a fidelity-based quality control mechanism filters them for consistency with user's approximate preference manifold, and a successor model is augmented on the enriched dataset. Our theoretical analysis shows that RSIR acts as a data-driven implicit regularizer, smoothing the optimization landscape and guiding models toward more robust solutions. Empirically, RSIR yields consistent, cumulative gains across multiple benchmarks and architectures. Notably, even smaller models benefit, and weak models can generate effective training curricula for stronger ones. These results demonstrate that recursive self-improvement is a general, model-agnostic approach to overcoming data sparsity, suggesting a scalable path forward for recommender systems and beyond. Our anonymized code is available at https://anonymous.4open.science/r/RSIR-7C5B .

Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control

TL;DR

The Recursive Self-Improving Recommendation framework is proposed, a paradigm in which a model bootstraps its own performance without reliance on external data or teacher models, suggesting a scalable path forward for recommender systems and beyond.

Abstract

The scarcity of high-quality training data presents a fundamental bottleneck to scaling machine learning models. This challenge is particularly acute in recommendation systems, where extreme sparsity in user interactions leads to rugged optimization landscapes and poor generalization. We propose the Recursive Self-Improving Recommendation (RSIR) framework, a paradigm in which a model bootstraps its own performance without reliance on external data or teacher models. RSIR operates in a closed loop: the current model generates plausible user interaction sequences, a fidelity-based quality control mechanism filters them for consistency with user's approximate preference manifold, and a successor model is augmented on the enriched dataset. Our theoretical analysis shows that RSIR acts as a data-driven implicit regularizer, smoothing the optimization landscape and guiding models toward more robust solutions. Empirically, RSIR yields consistent, cumulative gains across multiple benchmarks and architectures. Notably, even smaller models benefit, and weak models can generate effective training curricula for stronger ones. These results demonstrate that recursive self-improvement is a general, model-agnostic approach to overcoming data sparsity, suggesting a scalable path forward for recommender systems and beyond. Our anonymized code is available at https://anonymous.4open.science/r/RSIR-7C5B .
Paper Structure (70 sections, 37 equations, 7 figures, 14 tables, 1 algorithm)

This paper contains 70 sections, 37 equations, 7 figures, 14 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. Sequential Recommendation.
  4. Self-Improving AI.
  5. Methodology
  6. The Iterative Self-Improvement Loop
  7. Principled Synthetic Sequence Generation
  8. Bounded Exploration via a Hybrid Candidate Pool.
  9. Fidelity-Based Quality Control
  10. Computational Complexity Analysis
  11. Discussion and Theoretical Analysis
  12. Implicit Regularization and Landscape Smoothing
  13. Error Analysis and Stability Guarantee
  14. Experiments
  15. Experimental Settings
  16. ...and 55 more sections

Figures (7)

  • Figure 1: Overview of the Recursive Self-Improving Recommendation (RSIR) Framework.
  • Figure 2: Hyperparameter sensitivity of RSIR on Amazon-Sport.
  • Figure 3: RSIR performance over recursive iterations (NDCG@10 and Recall@10) on Amazon-Sport and Yelp.
  • Figure 4: Improvement Rate Heatmap for Weak-to-Strong Transfer in RSIR with a Student Model Trained on Synthetic Data from Teacher Models of Varying Strengths.
  • Figure 5: Analysis of Generated Data.
  • ...and 2 more figures

Theorems & Definitions (1)

  • proof