{"total":151,"rows":[{"id":"2407.08608","title":"FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision","created":"2024-07-11","cited_by_count":519,"area":null},{"id":"2212.09720","title":"The case for 4-bit precision: k-bit Inference Scaling Laws","created":"2022-12-19","cited_by_count":251,"area":null},{"id":"2411.04330","title":"Scaling Laws for Precision","created":"2024-11-07","cited_by_count":95,"area":null},{"id":"2403.14123","title":"AI and Memory Wall","created":"2024-03-21","cited_by_count":92,"area":null},{"id":"2302.14017","title":"Full Stack Optimization of Transformer Inference: a Survey","created":"2023-02-27","cited_by_count":80,"area":null},{"id":"2310.18313","title":"FP8-LM: Training FP8 Large Language Models","created":"2023-10-27","cited_by_count":77,"area":null},{"id":"2505.11594","title":"SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training","created":"2025-05-16","cited_by_count":66,"area":null},{"id":"2409.12517","title":"Scaling FP8 training to trillion-token LLMs","created":"2024-09-19","cited_by_count":65,"area":null},{"id":"2302.08007","title":"With Shared Microexponents, A Little Shifting Goes a Long Way","created":"2023-02-16","cited_by_count":50,"area":null},{"id":"2307.09782","title":"ZeroQuant-FP: A Leap Forward in LLMs Post-Training W4A8 Quantization Using Floating-Point Formats","created":"2023-07-19","cited_by_count":44,"area":null},{"id":"2407.08296","title":"Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients","created":"2024-07-11","cited_by_count":38,"area":null},{"id":"2410.19313","title":"COAT: Compressing Optimizer states and Activation for Memory-Efficient FP8 Training","created":"2024-10-25","cited_by_count":33,"area":null},{"id":"2303.17951","title":"FP8 versus INT8 for efficient deep learning inference","created":"2023-03-31","cited_by_count":31,"area":null},{"id":"2407.17465","title":"u-$\\mu$P: The Unit-Scaled Maximal Update Parametrization","created":"2024-07-24","cited_by_count":29,"area":null},{"id":"2406.12930","title":"Tender: Accelerating Large Language Models via Tensor Decomposition and Runtime Requantization","created":"2024-06-16","cited_by_count":28,"area":null},{"id":"2502.20853","title":"Oscillation-Reduced MXFP4 Training for Vision Transformers","created":"2025-02-28","cited_by_count":21,"area":null},{"id":"2601.22813","title":"Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation","created":"2026-01-30","cited_by_count":21,"area":null},{"id":"2306.02781","title":"An Automated Survey of Generative Artificial Intelligence: Large Language Models, Architectures, Protocols, and Applications","created":"2023-06-05","cited_by_count":20,"area":null},{"id":"2309.17224","title":"Training and inference of large language models using 8-bit floating point","created":"2023-09-29","cited_by_count":20,"area":null},{"id":"2305.12356","title":"Integer or Floating Point? New Outlooks for Low-Bit Quantization on Large Language Models","created":"2023-05-21","cited_by_count":19,"area":null},{"id":"2405.20935","title":"Effective Interplay between Sparsity and Quantization: From Theory to Practice","created":"2024-05-31","cited_by_count":19,"area":null},{"id":"2405.03103","title":"Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs","created":"2024-05-06","cited_by_count":18,"area":null},{"id":"2303.11257","title":"Unit Scaling: Out-of-the-Box Low-Precision Training","created":"2023-03-20","cited_by_count":17,"area":null},{"id":"2306.04039","title":"Revisiting Neural Retrieval on Accelerators","created":"2023-06-06","cited_by_count":14,"area":null},{"id":"2312.08583","title":"ZeroQuant(4+2): Redefining LLMs Quantization with a New FP6-Centric Strategy for Diverse Generative Tasks","created":"2023-12-14","cited_by_count":14,"area":null},{"id":"2409.11055","title":"Exploring the Trade-Offs: Quantization Methods, Task Difficulty, and Model Size in Large Language Models From Edge to Giant","created":"2024-09-17","cited_by_count":14,"area":null},{"id":"2503.08040","title":"Accurate INT8 Training Through Dynamic Block-Level Fallback","created":"2025-03-11","cited_by_count":14,"area":null},{"id":"2410.09615","title":"SLiM: One-shot Quantization and Sparsity with Low-rank Approximation for LLM Weight Compression","created":"2024-10-12","cited_by_count":12,"area":null},{"id":"2505.01043","title":"Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities","created":"2025-05-02","cited_by_count":11,"area":null},{"id":"2505.20524","title":"Towards Fully FP8 GEMM LLM Training at Scale","created":"2025-05-26","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":"2307.15034","title":"Guaranteed Approximation Bounds for Mixed-Precision Neural Operators","created":"2023-07-27","cited_by_count":9,"area":null},{"id":"2409.16997","title":"INT-FlashAttention: Enabling Flash Attention for INT8 Quantization","created":"2024-09-25","cited_by_count":9,"area":null},{"id":"2501.10658","title":"LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator","created":"2025-01-18","cited_by_count":9,"area":null},{"id":"2505.02309","title":"Optimizing LLMs for Resource-Constrained Environments: A Survey of Model Compression Techniques","created":"2025-05-05","cited_by_count":9,"area":null},{"id":"2509.00404","title":"Metis: Training LLMs with FP4 Quantization","created":"2025-08-30","cited_by_count":9,"area":null},{"id":"2501.02423","title":"Scaling Laws for Floating Point Quantization Training","created":"2025-01-05","cited_by_count":8,"area":null},{"id":"2607.04422","title":"Full-Stack FP4: Stable LLM Pretraining with Quantized Projections, Optimizers, and Attention","created":"2026-07-05","cited_by_count":8,"area":null},{"id":"2409.16626","title":"Ascend HiFloat8 Format for Deep Learning","created":"2024-09-25","cited_by_count":7,"area":null},{"id":"2501.04052","title":"RaZeR: Pushing the Limits of NVFP4 Quantization with Redundant Zero Remapping","created":"2025-01-06","cited_by_count":7,"area":null},{"id":"2605.18733","title":"Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory","created":"2026-05-18","cited_by_count":7,"area":null},{"id":"2308.03290","title":"FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search","created":"2023-08-07","cited_by_count":6,"area":null},{"id":"2407.00110","title":"Chat AI: A Seamless Slurm-Native Solution for HPC-Based Services","created":"2024-06-27","cited_by_count":6,"area":null},{"id":"2509.13765","title":"TENET: An Efficient Sparsity-Aware LUT-Centric Architecture for Ternary LLM Inference On Edge","created":"2025-09-17","cited_by_count":6,"area":null},{"id":"2502.05967","title":"$\\mu$nit Scaling: Simple and Scalable FP8 LLM Training","created":"2025-02-09","cited_by_count":5,"area":null},{"id":"2510.01650","title":"The Unseen Frontier: Pushing the Limits of LLM Sparsity with Surrogate-Free ADMM","created":"2025-10-02","cited_by_count":5,"area":null},{"id":"2510.02676","title":"To Compress or Not? Pushing the Frontier of Lossless GenAI Model Weights Compression with Exponent Concentration","created":"2025-10-03","cited_by_count":5,"area":null},{"id":"2510.16805","title":"Mixed-Precision Quantization for Language Models: Techniques and Prospects","created":"2025-10-19","cited_by_count":5,"area":null},{"id":"2602.23349","title":"FlashOptim: Optimizers for Memory-Efficient Training","created":"2026-02-26","cited_by_count":5,"area":null},{"id":"2604.15416","title":"StoSignSGD: Unbiased Structural Stochasticity Fixes SignSGD for Training Large Language Models","created":"2026-04-16","cited_by_count":5,"area":null},{"id":"2301.03904","title":"RedMule: A Mixed-Precision Matrix-Matrix Operation Engine for Flexible and Energy-Efficient On-Chip Linear Algebra and TinyML Training Acceleration","created":"2023-01-10","cited_by_count":4,"area":null},{"id":"2311.12359","title":"Shedding the Bits: Pushing the Boundaries of Quantization with Minifloats on FPGAs","created":"2023-11-21","cited_by_count":4,"area":null},{"id":"2312.02210","title":"Low-Precision Mixed-Computation Models for Inference on Edge","created":"2023-12-03","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":"2407.17353","title":"Scalify: scale propagation for efficient low-precision LLM training","created":"2024-07-24","cited_by_count":4,"area":null},{"id":"2506.20752","title":"Characterization and Mitigation of Training Instabilities in Microscaling Formats","created":"2025-06-25","cited_by_count":4,"area":null},{"id":"2601.06118","title":"Beyond Reproducibility: Token Probabilities Expose Large Language Model Nondeterminism","created":"2026-01-03","cited_by_count":4,"area":null},{"id":"2602.10718","title":"SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining","created":"2026-02-11","cited_by_count":4,"area":null},{"id":"2605.25880","title":"The Quantization Benefits of Residual-Free Transformers","created":"2026-05-25","cited_by_count":4,"area":null},{"id":"2606.11780","title":"What Limits Does Quantization Place on Dense Top-$k$ Retrieval? A Theoretical Study","created":"2026-06-10","cited_by_count":4,"area":null},{"id":"2211.10737","title":"Accuracy Booster: Enabling 4-bit Fixed-point Arithmetic for DNN Training","created":"2022-11-19","cited_by_count":3,"area":null},{"id":"2311.11172","title":"Low-Precision Floating-Point for Efficient On-Board Deep Neural Network Processing","created":"2023-11-18","cited_by_count":3,"area":null},{"id":"2511.06838","title":"P3-LLM: An Integrated NPU-PIM Accelerator for Edge LLM Inference Using Hybrid Numerical Formats","created":"2025-11-10","cited_by_count":3,"area":null},{"id":"2511.21513","title":"IntAttention: A Fully Integer Attention Pipeline for Efficient Edge Inference","created":"2025-11-26","cited_by_count":3,"area":null},{"id":"2604.03298","title":"ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs","created":"2026-03-28","cited_by_count":3,"area":null},{"id":"2605.12464","title":"Search Your Block Floating Point Scales!","created":"2026-05-12","cited_by_count":3,"area":null},{"id":"2605.13907","title":"AIS: Adaptive Importance Sampling for Quantized RL","created":"2026-05-13","cited_by_count":3,"area":null},{"id":"2606.04115","title":"dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formats","created":"2026-06-02","cited_by_count":3,"area":null},{"id":"2301.13464","title":"Training with Mixed-Precision Floating-Point Assignments","created":"2023-01-31","cited_by_count":2,"area":null},{"id":"2408.14528","title":"Adaptive Resolution Inference (ARI): Energy-Efficient Machine Learning for Internet of Things","created":"2024-08-26","cited_by_count":2,"area":null},{"id":"2503.21261","title":"HOT: Hadamard-based Optimized Training","created":"2025-03-27","cited_by_count":2,"area":null},{"id":"2505.14201","title":"FLASH-D: FlashAttention with Hidden Softmax Division","created":"2025-05-20","cited_by_count":2,"area":null},{"id":"2506.17768","title":"Log-Normal Multiplicative Dynamics for Stable Low-Precision Training of Large Networks","created":"2025-06-21","cited_by_count":2,"area":null},{"id":"2508.08822","title":"OISMA: On-the-fly In-memory Stochastic Multiplication Architecture for Matrix-Multiplication Workloads","created":"2025-08-12","cited_by_count":2,"area":null},{"id":"2511.05187","title":"Gradient Prediction with Control Variates in the Cheap-Forward Regime","created":"2025-11-07","cited_by_count":2,"area":null},{"id":"2511.05811","title":"MOSS: Efficient and Accurate FP8 LLM Training with Microscaling and Automatic Scaling","created":"2025-11-08","cited_by_count":2,"area":null},{"id":"2511.15300","title":"Quant-Trim in Practice: Improved Cross-Platform Low-Bit Deployment on Edge NPUs","created":"2025-11-19","cited_by_count":2,"area":null},{"id":"2512.12131","title":"BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models","created":"2025-12-13","cited_by_count":2,"area":null},{"id":"2512.15306","title":"LLMQ: Efficient Lower-Precision Pretraining for Consumer GPUs","created":"2025-12-17","cited_by_count":2,"area":null},{"id":"2601.16450","title":"On the Expressive Power of Floating-Point Transformers","created":"2026-01-23","cited_by_count":2,"area":null},{"id":"2602.18851","title":"Rank-Aware Spectral Bounds on Attention Logits for Stable Low-Precision Training","created":"2026-02-21","cited_by_count":2,"area":null},{"id":"2609.02846","title":"UE5M3 FP4 Block Scaling for Stable Language Model Pretraining","created":"2026-09-02","cited_by_count":2,"area":null},{"id":"2307.03712","title":"INT-FP-QSim: Mixed Precision and Formats For Large Language Models and Vision Transformers","created":"2023-07-07","cited_by_count":1,"area":null},{"id":"2309.12381","title":"Memory Efficient Mixed-Precision Optimizers","created":"2023-09-21","cited_by_count":1,"area":null},{"id":"2311.05034","title":"Just-in-time Quantization with Processing-In-Memory for Efficient ML Training","created":"2023-11-08","cited_by_count":1,"area":null},{"id":"2405.19284","title":"Optimizing Foundation Model Inference on a Many-tiny-core Open-source RISC-V Platform","created":"2024-05-29","cited_by_count":1,"area":null},{"id":"2504.21553","title":"Precision Where It Matters: A Novel Spike Aware Mixed-Precision Quantization Strategy for LLaMA-based Language Models","created":"2025-04-30","cited_by_count":1,"area":null},{"id":"2505.14638","title":"Dual Precision Quantization for Efficient and Accurate Deep Neural Networks Inference","created":"2025-05-20","cited_by_count":1,"area":null},{"id":"2506.09280","title":"TTrace: Lightweight Error Checking and Diagnosis for Distributed Training","created":"2025-06-10","cited_by_count":1,"area":null},{"id":"2508.21394","title":"AI Compute Architecture and Evolution Trends","created":"2025-08-29","cited_by_count":1,"area":null},{"id":"2510.24061","title":"FALQON: Accelerating LoRA Fine-tuning with Low-Bit Floating-Point Arithmetic","created":"2025-10-28","cited_by_count":1,"area":null},{"id":"2602.02707","title":"Every Bit Counts: A Theoretical Study of Precision-Expressivity Tradeoffs in Quantized Transformers","created":"2026-02-02","cited_by_count":1,"area":null},{"id":"2603.10444","title":"The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training","created":"2026-03-11","cited_by_count":1,"area":null},{"id":"2603.22324","title":"DAQ: Delta-Aware Quantization for Post-Training LLM Weight Compression","created":"2026-03-20","cited_by_count":1,"area":null},{"id":"2605.05331","title":"ViTok-v2: Scaling Native Resolution Auto-Encoders to 5 Billion Parameters","created":"2026-05-06","cited_by_count":1,"area":null},{"id":"2605.11111","title":"ShardTensor: Domain Parallelism for Scientific Machine Learning","created":"2026-05-11","cited_by_count":1,"area":null},{"id":"2605.11354","title":"Lite3R: A Model-Agnostic Framework for Efficient Feed-Forward 3D Reconstruction","created":"2026-05-12","cited_by_count":1,"area":null},{"id":"2605.20402","title":"Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor","created":"2026-05-19","cited_by_count":1,"area":null},{"id":"2605.28704","title":"Expressive Power of Floating-Point Neural Networks with Arbitrary Reduction Orders and Inexact Activation Implementations","created":"2026-05-27","cited_by_count":1,"area":null},{"id":"2606.04028","title":"Novel Aspects of IEEE SA P3109 Arithmetic Formats for Machine Learning","created":"2026-06-01","cited_by_count":1,"area":null}],"next_cursor":"100"}