{"total":92,"rows":[{"id":"2205.14135","title":"FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness","created":"2022-05-27","cited_by_count":3673,"area":null},{"id":"2207.00032","title":"DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale","created":"2022-06-30","cited_by_count":407,"area":null},{"id":"2310.17157","title":"Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time","created":"2023-10-26","cited_by_count":252,"area":null},{"id":"2304.12210","title":"A Cookbook of Self-Supervised Learning","created":"2023-04-24","cited_by_count":250,"area":null},{"id":"2305.07027","title":"EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attention","created":"2023-05-11","cited_by_count":172,"area":null},{"id":"2310.18313","title":"FP8-LM: Training FP8 Large Language Models","created":"2023-10-27","cited_by_count":77,"area":null},{"id":"2305.15805","title":"Dynamic Context Pruning for Efficient and Interpretable Autoregressive Transformers","created":"2023-05-25","cited_by_count":67,"area":null},{"id":"2401.16456","title":"SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design","created":"2024-01-29","cited_by_count":56,"area":null},{"id":"2403.07648","title":"Characterization of Large Language Model Development in the Datacenter","created":"2024-03-12","cited_by_count":46,"area":null},{"id":"2107.06419","title":"FLAT: An Optimized Dataflow for Mitigating Attention Bottlenecks","created":"2021-07-13","cited_by_count":39,"area":null},{"id":"2302.04869","title":"Reversible Vision Transformers","created":"2023-02-09","cited_by_count":33,"area":null},{"id":"2206.08898","title":"SimA: Simple Softmax-free Attention for Vision Transformers","created":"2022-06-17","cited_by_count":32,"area":null},{"id":"2203.12758","title":"Mokey: Enabling Narrow Fixed-Point Inference for Out-of-the-Box Floating-Point Transformer Models","created":"2022-03-23","cited_by_count":19,"area":null},{"id":"2405.07518","title":"SambaNova SN40L: Scaling the AI Memory Wall with Dataflow and Composition of Experts","created":"2024-05-13","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":"2310.03150","title":"Federated Fine-Tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly","created":"2023-10-04","cited_by_count":17,"area":null},{"id":"2405.10480","title":"Lean Attention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers","created":"2024-05-17","cited_by_count":16,"area":null},{"id":"2101.08734","title":"Clairvoyant Prefetching for Distributed Machine Learning I/O","created":"2021-01-21","cited_by_count":14,"area":null},{"id":"2106.04718","title":"FastSeq: Make Sequence Generation Faster","created":"2021-06-08","cited_by_count":13,"area":null},{"id":"2310.02065","title":"VENOM: A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores","created":"2023-10-03","cited_by_count":13,"area":null},{"id":"2106.16006","title":"Improving the Efficiency of Transformers for Resource-Constrained Devices","created":"2021-06-30","cited_by_count":12,"area":null},{"id":"2208.10895","title":"A Comprehensive Study of Real-Time Object Detection Networks Across Multiple Domains: A Survey","created":"2022-08-23","cited_by_count":12,"area":null},{"id":"2404.13528","title":"SmartMem: Layout Transformation Elimination and Adaptation for Efficient DNN Execution on Mobile","created":"2024-04-21","cited_by_count":12,"area":null},{"id":"2301.13062","title":"Operator Fusion in XLA: Analysis and Evaluation","created":"2023-01-30","cited_by_count":9,"area":null},{"id":"2407.14645","title":"Performance Modeling and Workload Analysis of Distributed Large Language Model Training and Inference","created":"2024-07-19","cited_by_count":9,"area":null},{"id":"2603.16952","title":"Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies","created":"2026-03-16","cited_by_count":9,"area":null},{"id":"2201.08683","title":"A Comprehensive Study of Vision Transformers on Dense Prediction Tasks","created":"2022-01-21","cited_by_count":8,"area":null},{"id":"2209.01346","title":"HammingMesh: A Network Topology for Large-Scale Deep Learning","created":"2022-09-03","cited_by_count":8,"area":null},{"id":"2309.15467","title":"Enabling Resource-efficient AIoT System with Cross-level Optimization: A survey","created":"2023-09-27","cited_by_count":8,"area":null},{"id":"2404.02945","title":"Optimizing the Deployment of Tiny Transformers on Low-Power MCUs","created":"2024-04-03","cited_by_count":8,"area":null},{"id":"2412.03317","title":"FlashAttention on a Napkin: A Diagrammatic Approach to Deep Learning IO-Awareness","created":"2024-12-04","cited_by_count":7,"area":null},{"id":"2208.13363","title":"Survey: Exploiting Data Redundancy for Optimization of Deep Learning","created":"2022-08-29","cited_by_count":6,"area":null},{"id":"2303.05295","title":"Dynamic Stashing Quantization for Efficient Transformer Training","created":"2023-03-09","cited_by_count":6,"area":null},{"id":"2507.01299","title":"La RoSA: Enhancing LLM Efficiency via Layerwise Rotated Sparse Activation","created":"2025-07-02","cited_by_count":6,"area":null},{"id":"2510.14719","title":"Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References","created":"2025-10-16","cited_by_count":6,"area":null},{"id":"2602.11374","title":"Retrieval-Aware Distillation for Transformer-SSM Hybrids","created":"2026-02-11","cited_by_count":6,"area":null},{"id":"2310.19991","title":"PolyThrottle: Energy-efficient Neural Network Inference on Edge Devices","created":"2023-10-30","cited_by_count":5,"area":null},{"id":"2404.07164","title":"PIM-Opt: Demystifying Distributed Optimization Algorithms on a Real-World Processing-In-Memory System","created":"2024-04-10","cited_by_count":5,"area":null},{"id":"2509.18886","title":"Confidential LLM Inference: Performance and Cost Across CPU and GPU TEEs","created":"2025-09-23","cited_by_count":5,"area":null},{"id":"2605.19269","title":"CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs","created":"2026-05-19","cited_by_count":5,"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":"2404.11788","title":"Understanding the Performance Horizon of the Latest ML Workloads with NonGEMM Workloads","created":"2024-04-17","cited_by_count":4,"area":null},{"id":"2104.08335","title":"Demystifying BERT: Implications for Accelerator Design","created":"2021-04-14","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":"2211.17246","title":"Pex: Memory-efficient Microcontroller Deep Learning through Partial Execution","created":"2022-11-30","cited_by_count":3,"area":null},{"id":"2402.13537","title":"EffLoc: Lightweight Vision Transformer for Efficient 6-DOF Camera Relocalization","created":"2024-02-21","cited_by_count":3,"area":null},{"id":"2407.19784","title":"Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting","created":"2024-07-29","cited_by_count":3,"area":null},{"id":"2410.17957","title":"MCUBERT: Memory-Efficient BERT Inference on Commodity Microcontrollers","created":"2024-10-23","cited_by_count":3,"area":null},{"id":"2411.01417","title":"BF-IMNA: A Bit Fluid In-Memory Neural Architecture for Neural Network Acceleration","created":"2024-11-03","cited_by_count":3,"area":null},{"id":"2601.16622","title":"E2Former-V2: On-the-Fly Equivariant Attention with Linear Activation Memory","created":"2026-01-23","cited_by_count":3,"area":null},{"id":"2201.11218","title":"DNNFuser: Generative Pre-Trained Transformer as a Generalized Mapper for Layer Fusion in DNN Accelerators","created":"2022-01-26","cited_by_count":2,"area":null},{"id":"2504.20401","title":"Nonlinear Computation with Linear Optics via Source-Position Encoding","created":"2025-04-29","cited_by_count":2,"area":null},{"id":"2304.07613","title":"STen: Productive and Efficient Sparsity in PyTorch","created":"2023-04-15","cited_by_count":1,"area":null},{"id":"2310.00574","title":"YFlows: Systematic Dataflow Exploration and Code Generation for Efficient Neural Network Inference using SIMD Architectures on CPUs","created":"2023-10-01","cited_by_count":1,"area":null},{"id":"2402.06173","title":"Comparison of parallel SMC and MCMC for Bayesian deep learning","created":"2024-02-09","cited_by_count":1,"area":null},{"id":"2409.17093","title":"BitQ: Tailoring Block Floating Point Precision for Improved DNN Efficiency on Resource-Constrained Devices","created":"2024-09-25","cited_by_count":1,"area":null},{"id":"2506.05300","title":"Power Law Guided Dynamic Sifting for Efficient Attention","created":"2025-06-05","cited_by_count":1,"area":null},{"id":"2510.25166","title":"A Study on Inference Latency for Vision Transformers on Mobile Devices","created":"2025-10-29","cited_by_count":1,"area":null},{"id":"2605.20799","title":"Instant GPU Efficiency Visibility at Fleet Scale","created":"2026-05-20","cited_by_count":1,"area":null},{"id":"2609.25405","title":"Efficient Iterative Retrieval with Heterogeneous Batching","created":"2026-09-21","cited_by_count":1,"area":null},{"id":"2104.05343","title":"An Efficient 2D Method for Training Super-Large Deep Learning Models","created":"2021-04-12","cited_by_count":0,"area":null},{"id":"2108.02572","title":"SINGA-Easy: An Easy-to-Use Framework for MultiModal Analysis","created":"2021-08-03","cited_by_count":0,"area":null},{"id":"2108.02725","title":"VisualTextRank: Unsupervised Graph-based Content Extraction for Automating Ad Text to Image Search","created":"2021-08-05","cited_by_count":0,"area":null},{"id":"2109.05778","title":"Text is NOT Enough: Integrating Visual Impressions into Open-domain Dialogue Generation","created":"2021-09-13","cited_by_count":0,"area":null},{"id":"2202.02830","title":"Discovering Personalized Semantics for Soft Attributes in Recommender Systems using Concept Activation Vectors","created":"2022-02-06","cited_by_count":0,"area":null},{"id":"2203.11011","title":"Reinforced MOOCs Concept Recommendation in Heterogeneous Information Networks","created":"2022-03-08","cited_by_count":0,"area":null},{"id":"2204.04043","title":"C-NMT: A Collaborative Inference Framework for Neural Machine Translation","created":"2022-04-08","cited_by_count":0,"area":null},{"id":"2205.08335","title":"Explanation-Guided Fairness Testing through Genetic Algorithm","created":"2022-05-16","cited_by_count":0,"area":null},{"id":"2207.14648","title":"Using Graph Neural Networks for Program Termination","created":"2022-07-28","cited_by_count":0,"area":null},{"id":"2301.01048","title":"A Theory of I/O-Efficient Sparse Neural Network Inference","created":"2023-01-03","cited_by_count":0,"area":null},{"id":"2301.02432","title":"Myths and Legends in High-Performance Computing","created":"2023-01-06","cited_by_count":0,"area":null},{"id":"2303.09664","title":"Tribe or Not? Critical Inspection of Group Differences Using TribalGram","created":"2023-03-16","cited_by_count":0,"area":null},{"id":"2307.03930","title":"Rosko: Row Skipping Outer Products for Sparse Matrix Multiplication Kernels","created":"2023-07-08","cited_by_count":0,"area":null},{"id":"2312.14441","title":"DMC4ML: Data Movement Complexity for Machine Learning","created":"2023-12-22","cited_by_count":0,"area":null},{"id":"2406.12478","title":"Accelerating Depthwise Separable Convolutions on Ultra-Low-Power Devices","created":"2024-06-18","cited_by_count":0,"area":null},{"id":"2407.15545","title":"Inverted Activations: Reducing Memory Footprint in Neural Network Training","created":"2024-07-22","cited_by_count":0,"area":null},{"id":"2411.17720","title":"MAS-Attention: Memory-Aware Stream Processing for Attention Acceleration on Resource-Constrained Edge Devices","created":"2024-11-20","cited_by_count":0,"area":null},{"id":"2504.03749","title":"Input Resolution Downsizing as a Compression Technique for Vision Deep Learning Systems","created":"2025-04-01","cited_by_count":0,"area":null},{"id":"2505.01637","title":"Morello: Compiling Fast Neural Networks with Dynamic Programming and Spatial Compression","created":"2025-05-03","cited_by_count":0,"area":null},{"id":"2508.10356","title":"Improving OCR for Historical Texts of Multiple Languages","created":"2025-08-14","cited_by_count":0,"area":null},{"id":"2510.05175","title":"Exact Causal Attention with 10% Fewer Operations","created":"2025-10-05","cited_by_count":0,"area":null},{"id":"2510.14614","title":"First Attentions Last: Better Exploiting First Attentions for Efficient Transformer Training","created":"2025-10-16","cited_by_count":0,"area":null},{"id":"2602.22274","title":"Positional-aware Spatio-Temporal Network for Large-Scale Traffic Prediction","created":"2026-02-25","cited_by_count":0,"area":null},{"id":"2604.09558","title":"VTC: DNN Compilation with Virtual Tensors for Data Movement Elimination","created":"2026-02-11","cited_by_count":0,"area":null},{"id":"2605.31500","title":"On Efficient Scaling of GNNs via IO-Aware Layers Implementations","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2607.21927","title":"RIS-Kernel: A Model-Agnostic Architecture for Long-Context LLM Inference via Sparse Attention","created":"2026-07-24","cited_by_count":0,"area":null},{"id":"2608.00029","title":"Nova: An End-to-End MLIR Compiler for Deep Learning","created":"2026-07-15","cited_by_count":0,"area":null},{"id":"2608.08961","title":"Gradient Under Microscope: Benchmarking Resource Utilization of Memory-Efficient Gradient Computation Methods","created":"2026-08-09","cited_by_count":0,"area":null},{"id":"2608.10805","title":"Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation","created":"2026-08-11","cited_by_count":0,"area":null},{"id":"2608.21555","title":"Tensor Seeks Layout: Formalizing Layout Selection for ML Compilers","created":"2026-08-21","cited_by_count":0,"area":null},{"id":"2608.24664","title":"Maia 200: A Software Defined Dataflow System for Large-scale AI Acceleration","created":"2026-08-25","cited_by_count":0,"area":null},{"id":"2609.32114","title":"Empowering Hybrid Attention Models on NPUs","created":"2026-09-26","cited_by_count":0,"area":null}],"next_cursor":null}