{"total":91,"rows":[{"id":"2309.06180","title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","created":"2023-09-12","cited_by_count":5944,"area":"inference"},{"id":"2104.04473","title":"Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM","created":"2021-04-09","cited_by_count":632,"area":null},{"id":"2104.00677","title":"Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis","created":"2021-04-01","cited_by_count":464,"area":null},{"id":"2104.07857","title":"ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning","created":"2021-04-16","cited_by_count":256,"area":null},{"id":"2111.10050","title":"Combined Scaling for Zero-shot Transfer Learning","created":"2021-11-19","cited_by_count":182,"area":null},{"id":"2403.14123","title":"AI and Memory Wall","created":"2024-03-21","cited_by_count":92,"area":null},{"id":"2104.05158","title":"Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models","created":"2021-04-12","cited_by_count":78,"area":null},{"id":"2102.07988","title":"TeraPipe: Token-Level Pipeline Parallelism for Training Large-Scale Language Models","created":"2021-02-16","cited_by_count":75,"area":null},{"id":"2104.14129","title":"ActNN: Reducing Training Memory Footprint via 2-Bit Activation Compressed Training","created":"2021-04-29","cited_by_count":61,"area":null},{"id":"2312.05516","title":"Stateful Large Language Model Serving with Pensieve","created":"2023-12-09","cited_by_count":61,"area":"inference"},{"id":"2105.11118","title":"Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads","created":"2021-05-24","cited_by_count":51,"area":null},{"id":"2204.12013","title":"Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs","created":"2022-04-26","cited_by_count":49,"area":null},{"id":"2210.00036","title":"Differentially Private Bias-Term Fine-tuning of Foundation Models","created":"2022-09-30","cited_by_count":48,"area":null},{"id":"2210.06640","title":"Compute-Efficient Deep Learning: Algorithmic Trends and Opportunities","created":"2022-10-13","cited_by_count":38,"area":null},{"id":"2206.11357","title":"GACT: Activation Compressed Training for Generic Network Architectures","created":"2022-06-22","cited_by_count":31,"area":null},{"id":"2107.10847","title":"Accelerating Quadratic Optimization with Reinforcement Learning","created":"2021-07-22","cited_by_count":30,"area":null},{"id":"2310.03294","title":"DISTFLASHATTN: Distributed Memory-efficient Attention for Long-context LLMs Training","created":"2023-10-05","cited_by_count":29,"area":null},{"id":"2210.09603","title":"Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programs","created":"2022-10-18","cited_by_count":28,"area":null},{"id":"2207.07697","title":"POET: Training Neural Networks on Tiny Devices with Integrated Rematerialization and Paging","created":"2022-07-15","cited_by_count":26,"area":null},{"id":"2107.01739","title":"KAISA: An Adaptive Second-Order Optimizer Framework for Deep Neural Networks","created":"2021-07-04","cited_by_count":20,"area":null},{"id":"2201.12023","title":"Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning","created":"2022-01-28","cited_by_count":20,"area":null},{"id":"2212.00824","title":"On-device Training: A First Overview on Existing Systems","created":"2022-12-01","cited_by_count":19,"area":null},{"id":"2111.06483","title":"Sequential Aggregation and Rematerialization: Distributed Full-batch Training of Graph Neural Networks on Large Graphs","created":"2021-11-11","cited_by_count":18,"area":null},{"id":"2307.09988","title":"TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge","created":"2023-07-19","cited_by_count":18,"area":null},{"id":"2110.10221","title":"The CoRa Tensor Compiler: Compilation for Ragged Tensors with Minimal Padding","created":"2021-10-19","cited_by_count":17,"area":null},{"id":"2401.00134","title":"Unicron: Economizing Self-Healing LLM Training at Scale","created":"2023-12-30","cited_by_count":16,"area":null},{"id":"2108.05818","title":"PatrickStar: Parallel Training of Pre-trained Models via Chunk-based Memory Management","created":"2021-08-12","cited_by_count":15,"area":null},{"id":"2211.05368","title":"A Comprehensive Survey on Distributed Training of Graph Neural Networks","created":"2022-11-10","cited_by_count":15,"area":null},{"id":"2408.08015","title":"Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices","created":"2024-08-15","cited_by_count":13,"area":null},{"id":"2311.03687","title":"Dissecting the Runtime Performance of the Training, Fine-tuning, and Inference of Large Language Models","created":"2023-11-07","cited_by_count":12,"area":null},{"id":"2206.04959","title":"Merak: An Efficient Distributed DNN Training Framework with Automated 3D Parallelism for Giant Foundation Models","created":"2022-06-10","cited_by_count":11,"area":null},{"id":"2302.08005","title":"Slapo: A Schedule Language for Progressive Optimization of Large Deep Learning Model Training","created":"2023-02-16","cited_by_count":11,"area":null},{"id":"2401.02643","title":"Training and Serving System of Foundation Models: A Comprehensive Survey","created":"2024-01-05","cited_by_count":11,"area":null},{"id":"2310.07147","title":"QFT: Quantized Full-parameter Tuning of LLMs with Affordable Resources","created":"2023-10-11","cited_by_count":10,"area":null},{"id":"2305.15265","title":"Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model","created":"2023-05-24","cited_by_count":9,"area":null},{"id":"2402.18789","title":"FlexLLM: Token-Level Co-Serving of LLM Inference and Finetuning with SLO Guarantees","created":"2024-02-29","cited_by_count":9,"area":null},{"id":"2407.02772","title":"Gradient descent with generalized Newton's method","created":"2024-07-03","cited_by_count":9,"area":null},{"id":"2411.13055","title":"Hardware Scaling Trends and Diminishing Returns in Large-Scale Distributed Training","created":"2024-11-20","cited_by_count":8,"area":null},{"id":"2503.19050","title":"Mist: Efficient Distributed Training of Large Language Models via Memory-Parallelism Co-Optimization","created":"2025-03-24","cited_by_count":7,"area":null},{"id":"2202.01306","title":"Harmony: Overcoming the Hurdles of GPU Memory Capacity to Train Massive DNN Models on Commodity Servers","created":"2022-02-02","cited_by_count":6,"area":null},{"id":"2206.14098","title":"RevBiFPN: The Fully Reversible Bidirectional Feature Pyramid Network","created":"2022-06-28","cited_by_count":6,"area":null},{"id":"2210.12924","title":"OLLA: Optimizing the Lifetime and Location of Arrays to Reduce the Memory Usage of Neural Networks","created":"2022-10-24","cited_by_count":6,"area":null},{"id":"2504.14519","title":"SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training","created":"2025-04-20","cited_by_count":6,"area":null},{"id":"2302.02599","title":"Colossal-Auto: Unified Automation of Parallelization and Activation Checkpoint for Large-scale Models","created":"2023-02-06","cited_by_count":5,"area":null},{"id":"2305.18513","title":"SlimFit: Memory-Efficient Fine-Tuning of Transformer-based Models Using Training Dynamics","created":"2023-05-29","cited_by_count":5,"area":null},{"id":"2403.01384","title":"On the Compressibility of Quantized Large Language Models","created":"2024-03-03","cited_by_count":5,"area":null},{"id":"2202.10435","title":"Survey on Large Scale Neural Network Training","created":"2022-02-21","cited_by_count":4,"area":null},{"id":"2204.02360","title":"Scientometric Review of Artificial Intelligence for Operations & Maintenance of Wind Turbines: The Past, Present and Future","created":"2022-03-30","cited_by_count":4,"area":null},{"id":"2210.10246","title":"Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction","created":"2022-10-19","cited_by_count":4,"area":null},{"id":"2307.01236","title":"Rockmate: an Efficient, Fast, Automatic and Generic Tool for Re-materialization in PyTorch","created":"2023-07-03","cited_by_count":4,"area":null},{"id":"2403.00176","title":"SoD$^2$: Statically Optimizing Dynamic Deep Neural Network","created":"2024-02-29","cited_by_count":4,"area":null},{"id":"2406.08334","title":"ProTrain: Efficient LLM Training via Memory-Aware Techniques","created":"2024-06-12","cited_by_count":4,"area":null},{"id":"2503.03777","title":"FlexInfer: Breaking Memory Constraint via Flexible and Efficient Offloading for On-Device LLM Inference","created":"2025-03-04","cited_by_count":4,"area":null},{"id":"2508.10395","title":"XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization","created":"2025-08-14","cited_by_count":4,"area":"inference"},{"id":"2110.08633","title":"Hydra: A System for Large Multi-Model Deep Learning","created":"2021-10-16","cited_by_count":3,"area":null},{"id":"2110.10802","title":"A Data-Centric Optimization Framework for Machine Learning","created":"2021-10-20","cited_by_count":3,"area":null},{"id":"2211.10948","title":"FedDCT: Federated Learning of Large Convolutional Neural Networks on Resource Constrained Devices using Divide and Collaborative Training","created":"2022-11-20","cited_by_count":3,"area":null},{"id":"2311.00591","title":"Coop: Memory is not a Commodity","created":"2023-11-01","cited_by_count":3,"area":null},{"id":"2506.06472","title":"Cost-Efficient LLM Training with Lifetime-Aware Tensor Offloading via GPUDirect Storage","created":"2025-06-06","cited_by_count":3,"area":null},{"id":"2203.15980","title":"DELTA: Dynamically Optimizing GPU Memory beyond Tensor Recomputation","created":"2022-03-30","cited_by_count":2,"area":null},{"id":"2209.10778","title":"Nesting Forward Automatic Differentiation for Memory-Efficient Deep Neural Network Training","created":"2022-09-22","cited_by_count":2,"area":null},{"id":"2211.16648","title":"COMET: A Comprehensive Cluster Design Methodology for Distributed Deep Learning Training","created":"2022-11-30","cited_by_count":2,"area":null},{"id":"2212.09290","title":"XEngine: Optimal Tensor Rematerialization for Neural Networks in Heterogeneous Environments","created":"2022-12-19","cited_by_count":2,"area":null},{"id":"2507.16274","title":"STAlloc: Enhancing Memory Efficiency in Large-Scale Model Training with Spatio-Temporal Planning","created":"2025-07-22","cited_by_count":2,"area":"moe"},{"id":"2201.09210","title":"Terra: Imperative-Symbolic Co-Execution of Imperative Deep Learning Programs","created":"2022-01-23","cited_by_count":1,"area":null},{"id":"2207.11912","title":"Dive into Big Model Training","created":"2022-07-25","cited_by_count":1,"area":null},{"id":"2303.04759","title":"RAF: Holistic Compilation for Deep Learning Model Training","created":"2023-03-08","cited_by_count":1,"area":null},{"id":"2307.09836","title":"Near-Linear Time Projection onto the $\\ell_{1,\\infty}$ Ball; Application to Sparse Autoencoders","created":"2023-07-19","cited_by_count":1,"area":"mechinterp"},{"id":"2402.14212","title":"Moonwalk: Inverse-Forward Differentiation","created":"2024-02-22","cited_by_count":1,"area":null},{"id":"2502.13811","title":"On the Duality between Gradient Transformations and Adapters","created":"2025-02-19","cited_by_count":1,"area":null},{"id":"2509.02197","title":"DaCe AD: Unifying High-Performance Automatic Differentiation for Machine Learning and Scientific Computing","created":"2025-09-02","cited_by_count":1,"area":null},{"id":"2510.11400","title":"FedHybrid: Breaking the Memory Wall of Federated Learning via Hybrid Tensor Management","created":"2025-10-13","cited_by_count":1,"area":null},{"id":"2604.27089","title":"AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism","created":"2026-04-29","cited_by_count":1,"area":null},{"id":"2607.02158","title":"Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs","created":"2026-07-02","cited_by_count":1,"area":null},{"id":"2609.31250","title":"Deterministic Regime Switching and Feasibility Inversion in Dynamic Tensor Rematerialization","created":"2026-09-25","cited_by_count":1,"area":null},{"id":"2303.14633","title":"An Evaluation of Memory Optimization Methods for Training Neural Networks","created":"2023-03-26","cited_by_count":0,"area":null},{"id":"2304.14463","title":"Moccasin: Efficient Tensor Rematerialization for Neural Networks","created":"2023-04-27","cited_by_count":0,"area":null},{"id":"2401.10652","title":"AutoChunk: Automated Activation Chunk for Memory-Efficient Long Sequence Inference","created":"2024-01-19","cited_by_count":0,"area":null},{"id":"2411.01611","title":"Stochastic Communication Avoidance for Recommendation Systems","created":"2024-11-03","cited_by_count":0,"area":null},{"id":"2506.08809","title":"Training-Free Inference for High-Resolution Sinogram Completion","created":"2025-06-10","cited_by_count":0,"area":null},{"id":"2509.26246","title":"SlimPack: Fine-Grained Asymmetric Packing for Balanced and Efficient Variable-Length LLM Training","created":"2025-09-30","cited_by_count":0,"area":null},{"id":"2512.17570","title":"GreedySnake: Accelerating SSD-Offloaded LLM Training with Efficient Scheduling and Optimizer Step Overlapping","created":"2025-12-19","cited_by_count":0,"area":null},{"id":"2601.19739","title":"TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching","created":"2026-01-27","cited_by_count":0,"area":null},{"id":"2603.15002","title":"MONET: Modeling and Optimization of neural NEtwork Training from Edge to Data Centers","created":"2026-03-16","cited_by_count":0,"area":null},{"id":"2606.08155","title":"Have I Solved This Before? Retrieving Similar Segmentation Problems for Evolutionary Learning","created":"2026-06-06","cited_by_count":0,"area":null},{"id":"2608.07524","title":"Training Variable Long Sequences with Data-Centric Parallel","created":"2026-07-14","cited_by_count":0,"area":null},{"id":"2608.08740","title":"Memory-Efficient Activation Checkpointing with Sliding Window and Hirschberg's Algorithm for 0/1 Knapsack Solving in PyTorch","created":"2026-08-09","cited_by_count":0,"area":null},{"id":"2609.14306","title":"Flattening Every Memory Peak in Long-Context Mixture-of-Experts Training","created":"2026-09-13","cited_by_count":0,"area":"moe"},{"id":"2609.24089","title":"FlashBoB: I/O-Efficient Exact Backward-over-Backward for Softmax Attention","created":"2026-09-21","cited_by_count":0,"area":null},{"id":"2609.27040","title":"EMA: Elastic and Performance Transparent Memory Across GPUs","created":"2026-09-22","cited_by_count":0,"area":null},{"id":"2610.14369","title":"When More Memory Fails: Hidden Feasibility Holes in Dynamic Tensor Rematerialization","created":"2026-10-09","cited_by_count":0,"area":null}],"next_cursor":null}