mlsanityData to 14 Oct 2026

Research on Model Parallelism and Data Parallelism Optimization Methods in Large Language Model-Based Recommendation Systems

arXiv 2506.17551 ↗First posted 21 Jun 2025cs.DC cs.AIcited by 6

Not on the map. Why?

Abstract (arXiv)

With the rapid adoption of large language models (LLMs) in recommendation systems, the computational and communication bottlenecks caused by their massive parameter sizes and large data volumes have become increasingly prominent. This paper systematically investigates two classes of optimization methods-model parallelism and data parallelism-for distributed training of LLMs in recommendation scenarios. For model parallelism, we implement both tensor parallelism and pipeline parallelism, and introduce an adaptive load-balancing mechanism to reduce cross-device communication overhead. For data parallelism, we compare synchronous and asynchronous modes, combining gradient compression and sparsification techniques with an efficient aggregation communication framework to significantly improve bandwidth utilization. Experiments conducted on a real-world recommendation dataset in a simulated service environment demonstrate that our proposed hybrid parallelism scheme increases training throughput by over 30% and improves resource utilization by approximately 20% compared to traditional single-mode parallelism, while maintaining strong scalability and robustness. Finally, we discuss trade-offs among different parallel strategies in online deployment and outline future directions involving heterogeneous hardware integration and automated scheduling technologies.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 20 of its 31 references that are in the map. First posted on arXiv 21 Jun 2025.

Builds on

20 of 31 references.

References · 20

  1. 2024Enhancing Code LLMs with Reinforcement Learning in Code Generation: A Surveycited by 83
  2. 2024A Multimodal Fusion Network For Student Emotion Recognition Based on Transformer and Tensor Productcited by 22
  3. 2024Research on Splicing Image Detection Algorithms Based on Natural Image Statistical Characteristicscited by 21
  4. 2024Enhanced Recommendation Combining Collaborative Filtering and Large Language Modelscited by 10
  5. 2024Real-Time Prediction for Athletes' Psychological States Using BERT-XGBoost: Enhancing Human-Computer Interactioncited by 9
  6. 2025Research on the Design of a Short Video Recommendation System Based on Multimodal Information and Differential Privacycited by 9
  7. 2025Multidimensional precipitation index prediction based on CNN-LSTM hybrid frameworkcited by 8
  8. 2025Personalized Risks and Regulatory Strategies of Large Language Models in Digital Advertisingcited by 8
  9. 2025User Behavior Analysis in Privacy Protection with Large Language Models: A Study on Privacy Preferences with Limited Datacited by 8
  10. 2024CCi-YOLOv8n: Enhanced Fire Detection with CARAFE and Context-Guided Modulescited by 7
  11. 2025Optimized Path Planning for Logistics Robots Using Ant Colony Algorithm under Multiple Constraintscited by 7
  12. 2025Psychological Health Knowledge-Enhanced LLM-based Social Network Crisis Intervention Text Transfer Recognition Methodcited by 7
8 more
  1. 2025II-NVM: Enhancing Map Accuracy and Consistency with Normal Vector-Assisted Mappingcited by 7
  2. 2025Avocado Price Prediction Using a Hybrid Deep Learning Model: TCN-MLP-Attention Architecturecited by 7
  3. 2025CTLformer: A Hybrid Denoising Model Combining Convolutional Layers and Self-Attention for Enhanced CT Image Reconstructioncited by 7
  4. 2025Research on Personalized Financial Product Recommendation by Integrating Large Language Models and Graph Neural Networkscited by 6
  5. 2025Research on E-Commerce Long-Tail Product Recommendation Mechanism Based on Large-Scale Language Modelscited by 6
  6. 2025Research on feature fusion and multimodal patent text based on graph attention networkcited by 5
  7. 2024Real-time Video Target Tracking Algorithm Utilizing Convolutional Neural Networks (CNN)cited by 4
  8. 2025Automated Parking Trajectory Generation Using Deep Reinforcement Learningcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 1.

6 papers.

  1. 2025CATCH: A Modular Cross-domain Adaptive Template with Hookcited by 6
  2. 2025Meta-Learning for Cold-Start Personalization in Prompt-Tuned LLMscited by 5
  3. 2025LLM-Augmented Symptom Analysis for Cardiovascular Disease Risk Prediction: A Clinical NLPcited by 2
  4. 2025Efficient Cold-Start Recommendation via BPE Token-Level Embedding Initialization with LLMcited by 2
  5. 2025Instructional Prompt Optimization for Few-Shot LLM-Based Recommendations on Cold-Start Userscited by 1
  6. 2026NestPipe: Large-Scale Recommendation Training on 1,500+ Accelerators via Nested Pipeliningcited by 0