Full-Precision Performance

Ask about this topic

Momentum

6 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28
View on map

Latest papers 42

All topics
CardsList

  1. Stochastic Rounding in Low-Precision Transformer Inference: A Variable-Precision Emulation Study of a Small GPT-2

    Oct 1, 2026Yohan Chatelain, Pablo de Oliveira CastroFull-Precision PerformanceLanguage Model Perplexity

    Krembil Centre for Neuroinformatics, CAMH, Toronto, Canada · Universite Paris-Saclay, UVSQ, LI-PaRAD, Versailles, France

  2. Backward-State Policy Is Part of the Learning Algorithm

    Sep 30, 2026Shuxiao Xie, Shuyang Xie, Dezhi Ran +2Backward PassesGradient

    Beijing Tongming Lake Information Technology Application Innovation Center (TLAIC), China

  3. DPS: Dual-Mode Precision LLM Serving with Semi-Unified Memory

    Sep 28, 2026Xuan Truong Nguyen, Tien Son Pham, Tuan Duc Chu +2Graphics Processing Unit MemoryKv-Cache Management

    Department of Next Generation Semiconductor Convergence and Open Sharing System (COSS), Seoul National University, Seoul 08826, South Korea · Van-Lang Institute of Semiconductor Technology (VIST), Hanoi, Vietnam · Efficient Computation Research Group, Moreh Vietnam +1

  4. The Undetected Damage of Quantization on Retrieval and How to Fix It

    Sep 21, 2026Luca Zhou, Alessandro Zirilli, Daniele Solombrino +2QuantizedFull-Precision Performance

    Sapienza University of Rome · Not Diamond · Paradigma

  5. Radio-Frequency Convolutional Neural Networks

    Sep 16, 2026Zhihui Gao, Shi-Yuan Ma, Yiran Chen +2Edge DevicesNeural Network Inference

    Department of Electrical and Computer Engineering, Duke University, Durham, NC 27708, USA. · Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA.

  6. Evaluating Losslessness in Speculative Decoding Under Finite-Precision Inference

    Sep 14, 2026Ilya Koziev, Leonid Sinev, Ivan OseledetsSpeculative DecodingFull-Precision Performance

  7. Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment

    Sep 2, 2026Anirudh Malik, M Sparsh Mehra, Poojith DevanQwen3Small Large Language Models

    OneBit AI

  8. CircuitsDNA: Discovering Unconventional Multi-Accuracy Arithmetic Circuits via Evolutionary Synthesis

    Sep 1, 2026Ruichen Qi, Junyi Luo, Xinting Jiang +5Field-Programmable Gate ArraysFull-Precision Performance

    Brown University · Google · University of Michigan

  9. In-Cell Learning: Language Models That Update Their Own Weights in Sequence Without Changing the File They Ship

    Aug 21, 2026Zifeng Liu, Yaxin Lu, Xuanhan Wu +6Model WeightsFull-Precision Performance

    Big Data and Artificial Intelligence Center, The Third Affiliated Hospital of Sun Yat-sen University · Institute for Frontier Interdisciplinary Research in Health Sciences and Technology, Sun Yat-sen University · Sun Yat-sen University Institute of Artificial Intelligence +7

  10. CurveFP: Co-Designing Numerical Representation and Product Arithmetic for Language Models

    Aug 8, 2026Ye QiaoFull-Precision PerformanceTask Arithmetic

    Department of Electrical Engineering and Computer Science University of California, Irvine Irvine, CA, USA

  11. HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models

    Jul 29, 2026Hei Yi Mak, Shadan Golestan, Hoang Le +10Fp8Full-Precision Performance

    Huawei

  12. When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation

    Jul 25, 2026Mojtaba SoltanalianFull-Precision PerformanceCircuit Depth

    Department of Electrical and Computer Engineering University of Illinois Chicago 851 S. Morgan St., MC 154 Chicago, IL 60607, USA

  13. Contraction-Gauge Preconditioning for Quantized Matrix Multiplication

    Jul 21, 2026Piyush Sao, Narasinga Miniskar, Pedro Valero-Lara +2Matrix MultiplicationSpectral Preconditioning

    Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA

  14. The Silent Freeze: Predicting When Low-Precision Training Stops Learning

    Jul 9, 2026Zekai ShangFull-Precision PerformanceFreeze

    University of Illinois at Urbana-Champaign, Champaign, IL, USA

  15. GeoFP8: Geometry-Aware FP8 Gradient Compression for Distributed LLM Training

    Jul 8, 2026Jieying Wang, Zizhong Wang, Fangru Linghu +3Large Language Model QuantizationLarge Language Model Pretraining

    Department of Computer Science, Rutgers University, Piscataway, NJ, USA · Department of Electrical and Computer Engineering, Rutgers University, Piscataway, NJ, USA

  16. Dithered Gaussian Mechanism for Randomness-Efficient Differential Privacy

    Jul 7, 2026Nikita P. Kalinin, Rasmus PaghStandard Differential-PrivacyRandom

    Institute of Science and Technology Austria · BARC University of Copenhagen

  17. How Should Transformers Encode Numeric Values in Electronic Health Records?

    Jul 1, 2026Maria Elkjær Montgomery, Christian Igel, Mikkel Odgaard +2Electronic Health RecordsTransformer Architectures

    Department of Computer Science, University of Copenhagen, Copenhagen, Denmark · Department of Organ Surgery and Transplantation, Copenhagen University Hospital - Rigshospitalet, Copenhagen, Denmark

  18. Extrapolating from Regularised Solutions for Solving Ill-Conditioned Linear Systems in Machine Learning

    Jun 29, 2026Disha Hegde, Jon Cockayne, Chris. J. OatesSolver IterationsRegularization

    University of Southampton, UK · Newcastle University, UK · The Alan Turing Institute, UK

  19. ReQAT: Achieving Full-Precision Reasoning Accuracy with 4-bit Floating-Point Quantization-Aware Training

    Jun 14, 2026Janghwan Lee, Sihwa Lee, Jinseok Kim +4Quantization-Aware TrainingFp8

    1Hanyang University, 2Rebellions Inc., Republic of Korea.

  20. Limitations of Learning Tanh Neural Networks with Finite Precision

    Jun 9, 2026Philipp Grohs, Matěj TrödlerActivation FunctionsNeural Network

    Faculty of Mathematics, University of Vienna · RICAM, Austrian Academy of Sciences

  21. dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formats

    Jun 2, 2026Giuseppe Franco, Ian Colbert, Pablo Monteagudo-Lago +2Mixed-Precision QuantizationFull-Precision Performance

    AMD

  22. P-Cast Precision in FP8 Attention: Sink-Induced Collapse and the Optimality of S=2^8

    Jun 2, 2026Reed LauFull-Precision PerformanceFp8

    Tencent

  23. Exact equivariance, kept through training, buys zero-shot generalisation across the symmetry group

    Jun 2, 2026Hongbo WangRotation EquivarianceEquivariance

    Department of Mathematics, Stony Brook University, Stony Brook, NY 11794, USA

  24. ScaleSweep: Accurate NVFP4 Post-Training Quantization of LLMs via Block Scale Initialization

    May 30, 2026Li Lin, Xiaojun WanFp8Full-Precision Performance

    Wangxuan Institute of Computer Technology, Peking University

  25. GNMR: Runtime Stability Control for Low-Precision Large Language Model Training

    May 30, 2026Boao Kong, Weichen Jia, Engao Zhang +6Large Language Model TrainingFull-Precision Performance

    Peking University · Huawei Technologies Ltd.

  26. Stochastic Rounding Increases Small Singular Values

    May 29, 2026Linkai Ma, Tingzhou Yu, Petros DrineasSingular Value DecompositionFull-Precision Performance

    Department of Mathematics University of Alberta

  27. Fast and Stable Triangular Inversion for Delta-Rule Linear Transformers

    May 20, 2026Aleksandros Sobczyk, Gioele Gottardo, Christos K. Matzoros +4Kimi Delta AttentionMatrix Multiplication

    Computing Systems Lab Huawei Technologies2026 Switzerland

  28. FTerViT: Fully Ternary Vision Transformer

    May 20, 2026Szymon Ruciński, Pietro Bonazzi, Engin Türetken +3Self-Supervised Vision TransformersVision Transformer

    CSEM, Neuchâtel, Switzerland · ETH Zürich, Zurich, Switzerland

  29. Exploring Bottlenecks in VLM-LLM Navigation: How 3D Scene Understanding Capability Impacts Zero-Shot VLN

    May 14, 2026Ziyi Xia, Chaoran Xiong, Litao Wei +2Vision-Language Navigation3D Scene Understanding

    Shanghai Key Laboratory of Navigation and Location Based Services, Shanghai Jiao Tong University (SJTU), Shanghai 200240, China · State Key Laboratory of Submarine Geoscience, SJTU

  30. Direction-Preserving Number Representations

    May 8, 2026Bardia Zadeh, George A. ConstantinidesFull-Precision PerformanceVectors

    Department of Electrical and Electronic Engineering Imperial College London London, United Kingdom

  31. EULER-ADAS: Energy-Efficient & SIMD-Unified Logarithmic-Posit Engine for Precision-Reconfigurable Approximate ADAS Acceleration

    May 7, 2026Mukul Lokhande, Ratko Pilipovic, Omkar Kokane +2Field-Programmable Gate ArraysEnergy-Efficient

    NSDCS Research Group, Dept. of Electrical Engineering, Indian Institute of Technology Indore, Indore, India · Faculty of Computer and Information Science, University of Ljubljana, Ljubljana, Slovenia · EnICS Labs, Bar-Ilan Univ., Ramat Gan 5290002, Israel

  32. Grokking or Glitching? How Low-Precision Drives Slingshot Loss Spikes

    May 7, 2026Liu Hanqing, Jianjun Cao, Yuanze Li +1Full-Precision PerformanceGrokking

    1Tsinghua University · 2The University of Tokyo

  33. Normalized Architectures are Natively 4-Bit

    May 7, 2026Maxim Fishman, Brian Chmiel, Ron Banner +2Large Language Model QuantizationFull-Precision Performance

    ∧NVIDIA · ◦Department of Electrical and Computer Engineering - Technion, Haifa, Israel

  34. Hardware-Aware Neural Feature Extraction for Resource-Constrained Devices

    May 5, 2026Francesco Tosini, Simone Pedroni, Christian Veronesi +5Resource-Constrained Edge DevicesFeature Extraction

    Department of Electronics, Information and Bioengineering (DEIB), Politecnico di Milano, Italy · Department of Electronics, Informationة · Smart Eyewear Lab, EssilorLuxottica, Milan, Italy

  35. DASH-KV: Accelerating Long-Context LLM Inference via Asymmetric KV Cache Hashing

    Apr 21, 2026Jinyu Guo, Zhihan Zhang, Jiehui Xie +7Key-Value Cache CompressionDepthweave-Kv

    School of Information and Software Engineering, University of Electronic Science and Technology of China · School of Computer Science and Engineering, University of Electronic Science and Technology of China · Bangladesh University of Engineering and Technology +2

  36. Training Time Prediction for Mixed Precision-based Distributed Training

    Apr 17, 2026Minchul Kang, Changyong Shin, Jinwoo Jeong +5Training TimeFull-Precision Performance

    Korea University · KT Corporation

  37. ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs

    Jan 12, 2026Haoqian Meng, Yilun Luo, Yafei Zhao +3Full-Precision PerformanceEfficient Large Language Model

    School of Computer Science and Technology, Tianjin University

  38. Precision autotuning for linear solvers via contextual bandit-based RL

    Jan 2, 2026Erin Carson, Xinye ChenSolver IterationsFull-Precision Performance

    Charles University Prague, Czech · Sorbonne Université, CNRS, LIP6 Paris, France

  39. Numerical stability analysis of large language models

    Mar 13, 2025Stanislav Budzinskiy, Wenyi Fang, Longbin Zeng +1Transformer ArchitecturesFull-Precision Performance

    Faculty of Mathematics, University of Vienna, Kolingasse 14-16, 1090, Vienna, Austria · Huawei Technologies

  40. Comparing YOLOv11 and YOLOv8 for instance segmentation of occluded and non-occluded immature green fruits in complex orchard environment

    Oct 24, 2024Ranjan Sapkota, Manoj KarkeeYolo26Full-Precision Performance

    Washington State University Department of Biological Systems Engineering

  41. Deep Network Approximation: Beyond ReLU to Diverse Activation Functions

    Jul 13, 2023Shijun Zhang, Jianfeng Lu, Hongkai ZhaoActivation FunctionsRectified Linear Unit Networks

    Department of Mathematics Duke University

  42. FP8 is All You Need (Part 1): Debunking Hardware FP64 as the HPC Holy Grail (Sep 3rd version)

    Date pendingSatoshi MatsuokaHigh-Performance ComputingFp8

    Director, RIKEN Center for Computational Science (R-CCS) Kobe, Hyogo, Japan