The field of recommender systems and language models is rapidly evolving, with a focus on improving personalization and reducing bias. Recent studies have explored the use of deep learning techniques, such as graph transformers and auto-distillation, to enhance the performance of recommender systems. Additionally, there is a growing interest in leveraging large language models to improve the accuracy and diversity of recommendations. Noteworthy papers in this area include those that propose novel frameworks for multimodal recommendation, such as MSCRS and HeLLM, which integrate multi-modal semantic graphs with prompt learning and hypergraph-based context. Other notable works focus on addressing the challenges of bias and fairness in recommender systems, such as the use of LLMs as prompt modifiers to avoid biases in AI image generators. Overall, the field is moving towards more sophisticated and nuanced approaches to recommendation, incorporating multiple modalities and contexts to provide more accurate and personalized suggestions. Noteworthy papers include: MSCRS, which proposes a multi-modal semantic graph prompt learning framework for conversational recommender systems. HeLLM, which introduces a novel framework for multimodal recommendation that equips LLMs with the capability to capture intricate higher-order semantic correlations.
Advancements in Recommender Systems and Language Models
Sources
On the Practice of Deep Hierarchical Ensemble Network for Ad Conversion Rate Prediction
Evaluating the Bias in LLMs for Surveying Opinion and Decision Making in Healthcare
A Comparative Study of Recommender Systems under Big Data Constraints
Leveraging Auto-Distillation and Generative Self-Supervised Learning in Residual Graph Transformers for Enhanced Recommender Systems
PinRec: Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems
JEPA4Rec: Learning Effective Language Representations for Sequential Recommendation via Joint Embedding Predictive Architecture
Distilling Transitional Pattern to Large Language Models for Multimodal Session-based Recommendation
Multi-Modal Hypergraph Enhanced LLM Learning for Recommendation
Integrating Textual Embeddings from Contrastive Learning with Generative Recommender for Enhanced Personalization
Epistemic Uncertainty-aware Recommendation Systems via Bayesian Deep Ensemble Learning
MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
Why am I seeing this? Towards recognizing social media recommender systems with missing recommendations
Using LLMs as prompt modifier to avoid biases in AI image generators
Cancer-Myth: Evaluating AI Chatbot on Patient Questions with False Presuppositions
Exploring the Impact of Personality Traits on Conversational Recommender Systems: A Simulation with Large Language Models
SimUSER: Simulating User Behavior with Large Language Models for Recommender System Evaluation
CSMF: Cascaded Selective Mask Fine-Tuning for Multi-Objective Embedding-Based Retrieval
Should We Tailor the Talk? Understanding the Impact of Conversational Styles on Preference Elicitation in Conversational Recommender Systems