Imagen de Portada

Superpipeline: A Universal Approach for Reducing GPU Memory Usage in Large Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Abbasi, Reza, Lim, Sernam
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

GPU Memory Usage Optimization for Backward Propagation in Deep Network Training
por: Hong, Ding-Yong, et al.
Publicado: (2025)

LLMem: Estimating GPU Memory Usage for Fine-Tuning Pre-Trained LLMs
por: Kim, Taeho, et al.
Publicado: (2024)

Tri-Accel: Curvature-Aware Precision-Adaptive and Memory-Elastic Optimization for Efficient GPU Usage
por: Sheibanian, Mohsen, et al.
Publicado: (2025)

On the Universality of Transformer Architectures; How Much Attention Is Enough?
por: Abbasi, Amirreza, et al.
Publicado: (2025)

ZO2: Scalable Zeroth-Order Fine-Tuning for Extremely Large Language Models with Limited GPU Memory
por: Wang, Liangyu, et al.
Publicado: (2025)

Opportunities in deep learning methods development for computational biology
por: Lee, Alex Jihun, et al.
Publicado: (2024)

Expanding Sparse Tuning for Low Memory Usage
por: Shen, Shufan, et al.
Publicado: (2024)

Reducing the GPU Memory Bottleneck with Lossless Compression for ML -- Extended
por: Kamath, Aditya K, et al.
Publicado: (2026)

Mastering Memory Tasks with World Models
por: Samsami, Mohammad Reza, et al.
Publicado: (2024)

GPU Memory Prediction for Multimodal Model Training
por: Jeong, Jinwoo, et al.
Publicado: (2025)

MeMoir: A Software-Driven Covert Channel based on Memory Usage
por: Gonzalez-Gomez, Jeferson, et al.
Publicado: (2024)

Mind the Memory Gap: Unveiling GPU Bottlenecks in Large-Batch LLM Inference
por: Recasens, Pol G., et al.
Publicado: (2025)

Loki: Representation over Architecture for Diffusion-Based Portrait Animation
por: Navard, Pouyan, et al.
Publicado: (2026)

FloE: On-the-Fly MoE Inference on Memory-constrained GPU
por: Zhou, Yuxin, et al.
Publicado: (2025)

Universality in Transfer Learning for Linear Models
por: Ghane, Reza, et al.
Publicado: (2024)

Physical Activation Functions (PAFs): An Approach for More Efficient Induction of Physics into Physics-Informed Neural Networks (PINNs)
por: Abbasi, Jassem, et al.
Publicado: (2022)

Scaling On-Device GPU Inference for Large Generative Models
por: Tang, Jiuqiang, et al.
Publicado: (2025)

Gaussian Universality for Diffusion Models
por: Ghane, Reza, et al.
Publicado: (2025)

GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks
por: Li, Zihao, et al.
Publicado: (2026)

Ready2Unlearn: A Learning-Time Approach for Preparing Models with Future Unlearning Readiness
por: Duan, Hanyu, et al.
Publicado: (2025)

Machine Learning Applications to Computational Plasma Physics and Reduced-Order Plasma Modeling: A Perspective
por: Faraji, Farbod, et al.
Publicado: (2024)

xMem: A CPU-Based Approach for Accurate Estimation of GPU Memory in Deep Learning Training Workloads
por: Shi, Jiabo, et al.
Publicado: (2025)

Inverted Activations: Reducing Memory Footprint in Neural Network Training
por: Novikov, Georgii, et al.
Publicado: (2024)

Reducing Fine-Tuning Memory Overhead by Approximate and Memory-Sharing Backpropagation
por: Yang, Yuchen, et al.
Publicado: (2024)

GPU-Accelerated INT8 Quantization for KV Cache Compression in Large Language Models
por: Taneja, Maanas, et al.
Publicado: (2026)

Reinforcement Learning for Dynamic Memory Allocation
por: Lim, Arisrei, et al.
Publicado: (2024)

Mobility-Assisted Decentralized Federated Learning: Convergence Analysis and A Data-Driven Approach
por: Jahani, Reza, et al.
Publicado: (2025)

Learning Latent Space Dynamics with Model-Form Uncertainties: A Stochastic Reduced-Order Modeling Approach
por: Yong, Jin Yi, et al.
Publicado: (2024)

LLMPerf: GPU Performance Modeling meets Large Language Models
por: Nguyen, Khoi N. M., et al.
Publicado: (2025)

Parallel Track Transformers: Enabling Fast GPU Inference with Reduced Synchronization
por: Wang, Chong, et al.
Publicado: (2026)

Multi-Stage Prototype Learning for Interpretable Time Series Classification
por: Kalisetti, Bhavesh, et al.
Publicado: (2021)

Memory-Reduced Meta-Learning with Guaranteed Convergence
por: Yang, Honglin, et al.
Publicado: (2024)

Semantic-Aware Scheduling for GPU Clusters with Large Language Models
por: Wang, Zerui, et al.
Publicado: (2025)

Online Learning Models for Vehicle Usage Prediction During COVID-19
por: Lindroth, Tobias, et al.
Publicado: (2022)

FRUGAL: Memory-Efficient Optimization by Reducing State Overhead for Scalable Training
por: Zmushko, Philip, et al.
Publicado: (2024)

Reducing Smoothness with Expressive Memory Enhanced Hierarchical Graph Neural Networks
por: Bailie, Thomas, et al.
Publicado: (2025)

The Mosaic Memory of Large Language Models
por: Shilov, Igor, et al.
Publicado: (2024)

SoftDTW-CUDA-Torch: Memory-Efficient GPU-Accelerated Soft Dynamic Time Warping for PyTorch
por: Weber, Ron Shapira, et al.
Publicado: (2026)

LLM Enhancer: Merged Approach using Vector Embedding for Reducing Large Language Model Hallucinations with External Knowledge
por: Rayhan, Naheed, et al.
Publicado: (2025)

PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU
por: Song, Yixin, et al.
Publicado: (2023)