Image de couverture de livre

The AetherFloat Family: Block-Scale-Free Quad-Radix Floating-Point Architectures for AI Accelerators

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Morisaki, Keita
Format:	Preprint
Publié:	2026
Sujets:	Hardware Architecture Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Search Your Block Floating Point Scales!
par: Gupta, Tanmaey, et autres
Publié: (2026)

Scaling Laws for Floating Point Quantization Training
par: Sun, Xingwu, et autres
Publié: (2025)

F-BFQ: Flexible Block Floating-Point Quantization Accelerator for LLMs
par: Haris, Jude, et autres
Publié: (2025)

SafeCiM: Investigating Resilience of Hybrid Floating-Point Compute-in-Memory Deep Learning Accelerators
par: Bhattacharya, Swastik, et autres
Publié: (2025)

MGS: Markov Greedy Sums for Accurate Low-Bitwidth Floating-Point Accumulation
par: Natesh, Vikas, et autres
Publié: (2025)

Schrödinger's FP: Dynamic Adaptation of Floating-Point Containers for Deep Learning Training
par: Nikolić, Miloš, et autres
Publié: (2022)

BBAL: A Bidirectional Block Floating Point-Based Quantisation Accelerator for Large Language Models
par: Han, Xiaomeng, et autres
Publié: (2025)

TimeFloats: Train-in-Memory with Time-Domain Floating-Point Scalar Products
par: Hashem, Maeesha Binte, et autres
Publié: (2024)

HiFloat4 Format for Language Model Inference
par: Luo, Yuanyong, et autres
Publié: (2026)

Ascend HiFloat8 Format for Deep Learning
par: Luo, Yuanyong, et autres
Publié: (2024)

AI Accelerators for Large Language Model Inference: Architecture Analysis and Scaling Strategies
par: Sharma, Amit
Publié: (2025)

A Hybrid-Domain Floating-Point Compute-in-Memory Architecture for Efficient Acceleration of High-Precision Deep Neural Networks
par: Yi, Zhiqiang, et autres
Publié: (2025)

MXFormer: A Microscaling Floating-Point Charge-Trap Transistor Compute-in-Memory Transformer Accelerator
par: Karfakis, George, et autres
Publié: (2026)

H-FA: A Hybrid Floating-Point and Logarithmic Approach to Hardware Accelerated FlashAttention
par: Alexandridis, Kosmas, et autres
Publié: (2025)

Range, Not Precision: Block-Floating-Point Half-Precision FFT and SAR Imaging on Apple Silicon
par: Bergach, Mohamed Amine
Publié: (2026)

Fast Generation of Custom Floating-Point Spatial Filters on FPGAs
par: Campos, Nelson, et autres
Publié: (2024)

Online Alignment and Addition in Multi-Term Floating-Point Adders
par: Alexandridis, Kosmas, et autres
Publié: (2024)

Floating Point HUB Adder for RISC-V Sargantana Processor
par: Bandera, Gerardo, et autres
Publié: (2024)

Efficient In-Memory Acceleration of Sparse Block Diagonal LLMs
par: de Lima, João Paulo Cardoso, et autres
Publié: (2025)

From Quarter to All: Accelerating Speculative LLM Decoding via Floating-Point Exponent Remapping and Parameter Sharing
par: Zhao, Yushu, et autres
Publié: (2025)

Accelerator-assisted Floating-point ASIP for Communication and Positioning in Massive MIMO Systems
par: Attari, Mohammad, et autres
Publié: (2025)

Converting Binary Floating-Point Numbers to Shortest Decimal Strings: An Experimental Review
par: Gareau, Jaël Champagne, et autres
Publié: (2026)

Floating-Point Multiply-Add with Approximate Normalization for Low-Cost Matrix Engines
par: Alexandridis, Kosmas, et autres
Publié: (2024)

Accelerating Computer Architecture Simulation through Machine Learning
par: Ali, Wajid, et autres
Publié: (2024)

Formal that "Floats" High: Formal Verification of Floating Point Arithmetic
par: Mohanty, Hansa, et autres
Publié: (2025)

A Stochastic Rounding-Enabled Low-Precision Floating-Point MAC for DNN Training
par: Ali, Sami Ben, et autres
Publié: (2024)

End-to-End Transformer Acceleration Through Processing-in-Memory Architectures
par: Yang, Xiaoxuan, et autres
Publié: (2025)

EVA: Accelerating LLM Decoding via an Efficient Vector Quantization Architecture
par: Duan, Bowen, et autres
Publié: (2026)

FLAASH: Flexible Accelerator Architecture for Sparse High-Order Tensor Contraction
par: Kulp, Gabriel, et autres
Publié: (2024)

TransDot: An Area-efficient Reconfigurable Floating-Point Unit for Trans-Precision Dot-Product Accumulation for FPGA AI Engines
par: Wang, Jiayi, et autres
Publié: (2026)

COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference
par: Qiao, Ye, et autres
Publié: (2025)

Atleus: Accelerating Transformers on the Edge Enabled by 3D Heterogeneous Manycore Architectures
par: Dhingra, Pratyush, et autres
Publié: (2025)

Inexactness and Correction of Floating-Point Reciprocal, Division and Square Root
par: Dutton, Lucas M., et autres
Publié: (2024)

Closing the Gap Between Float and Posit Hardware Efficiency
par: Jonnalagadda, Aditya Anirudh, et autres
Publié: (2026)

A Hybrid Residue Floating Numerical Architecture with Formal Error Bounds for High Throughput FPGA Computation
par: Darvishi, Mostafa
Publié: (2026)

Toward Attention-based TinyML: A Heterogeneous Accelerated Architecture and Automated Deployment Flow
par: Wiese, Philip, et autres
Publié: (2024)

E2AFS: Energy-Efficient Approximate Floating Point Square Rooter for Error Tolerant Computing
par: Goyal, Prateek, et autres
Publié: (2026)

MXDOTP: A RISC-V ISA Extension for Enabling Microscaling (MX) Floating-Point Dot Products
par: İslamoğlu, Gamze, et autres
Publié: (2025)

Accelerating Diffusion Models for Generative AI Applications with Silicon Photonics
par: Suresh, Tharini, et autres
Publié: (2026)

Memory Is All You Need: An Overview of Compute-in-Memory Architectures for Accelerating Large Language Model Inference
par: Wolters, Christopher, et autres
Publié: (2024)