Advances in Recommender Systems

The field of recommender systems is moving towards addressing the challenges of group decision-making, popularity bias, and cold start problems. Researchers are exploring innovative approaches to improve recommendation accuracy, fairness, and user satisfaction. Notably, there is a growing interest in developing context-aware and multi-criteria group recommender systems, as well as exposure-aware retrieval scoring approaches to mitigate popularity bias. Additionally, novel token parameterization techniques, such as Semantic ID prefix ngram, are being proposed to enhance embedding representation stability. Furthermore, content-based filtering methods are being improved by considering users' information seeking behaviors. Some noteworthy papers in this area include: Finding Interest Needle in Popularity Haystack, which introduces an exposure-aware retrieval scoring approach to mitigate popularity bias. Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID, which proposes a novel token parameterization technique to improve embedding representation stability. FEASE: Shallow AutoEncoding Recommender with Cold Start Handling via Side Features, which presents a straightforward, autoencoder-based method to address cold start issues.

Sources

From Individual to Group: Developing a Context-Aware Multi-Criteria Group Recommender System

Reproducibility Companion Paper: Making Users Indistinguishable: Attribute-wise Unlearning in Recommender Systems

Reproducibility Companion Paper:In-processing User Constrained Dominant Sets for User-Oriented Fairness in Recommender Systems

Finding Interest Needle in Popularity Haystack: Improving Retrieval by Modeling Item Exposure

Contextual Preference Collaborative Measure Framework Based on Belief System

Extending MovieLens-32M to Provide New Evaluation Objectives

Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID

FEASE: Shallow AutoEncoding Recommender with Cold Start Handling via Side Features

Research Paper Recommender System by Considering Users' Information Seeking Behaviors

Built with on top of