SOTAVerified

Quantization

Quantization is a promising technique to reduce the computation cost of neural network training, which can replace high-cost floating-point numbers (e.g., float32) with low-cost fixed-point numbers (e.g., int8/int16).

Source: Adaptive Precision Training: Quantify Back Propagation in Neural Networks with Fixed-point Numbers

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1951–2000 of 4925 papers

TitleStatusHype
FAT: An In-Memory Accelerator with Fast Addition for Ternary Weight Neural Networks—0
Fast top-K Cosine Similarity Search through XOR-Friendly Binary Quantization on GPUs—0
Compact and Robust Deep Learning Architecture for Fluorescence Lifetime Imaging and FPGA Implementation—0
Generative Design of Hardware-aware DNNs—0
Generative Diffusion Models for Lattice Field Theory—0
Fast Template Evaluation with Vector Quantization—0
Generative QoE Modeling: A Lightweight Approach for Telecom Networks—0
Generative Semantic Communication for Text-to-Speech Synthesis—0
A Reconstruction-Computation-Quantization (RCQ) Approach to Node Operations in LDPC Decoding—0
A Deep Hashing Learning Network—0
FastSGD: A Fast Compressed SGD Framework for Distributed Machine Learning—0
FastQuery: Communication-efficient Embedding Table Query for Private LLM Inference—0
A reconfigurable neural network ASIC for detector front-end data compression at the HL-LHC—0
Geometry and clustering with metrics derived from separable Bregman divergences—0
Fast Orthogonal Projection Based on Kronecker Product—0
Getting Free Bits Back from Rotational Symmetries in LLMs—0
Fast on-line signature recognition based on VQ with time modeling—0
GHN-Q: Parameter Prediction for Unseen Quantized Convolutional Architectures via Graph Hypernetworks—0
A Reconfigurable Dual-Mode Tracking SAR ADC without Analog Subtraction—0
Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions—0
3D Pathfinding and Collision Avoidance Using Uneven Search-space Quantization and Visual Cone Search—0
Givens Coordinate Descent Methods for Rotation Matrix Learning in Trainable Embedding Indexes—0
Synaptic Modulation using Interspike Intervals Increases Energy Efficiency of Spiking Neural Networks—0
FastMamba: A High-Speed and Efficient Mamba Accelerator on FPGA with Accurate Quantization—0
Communication-efficient Variance-reduced Stochastic Gradient Descent—0
Global synchronization of multi-agent systems with nonlinear interactions—0
Goal-oriented compression for L_p-norm-type goal functions: Application to power consumption scheduling—0
Goal-Oriented Quantization: Analysis, Design, and Application to Resource Allocation—0
Fast Low-rank Representation based Spatial Pyramid Matching for Image Classification—0
GOBO: Quantizing Attention-Based NLP Models for Low Latency and Energy Efficient Inference—0
Communication-Efficient Split Learning via Adaptive Feature-Wise Compression—0
Going Further With Winograd Convolutions: Tap-Wise Quantization for Efficient Inference on 4x4 Tile—0
Google USM: Scaling Automatic Speech Recognition Beyond 100 Languages—0
Fast learning rates with heavy-tailed losses—0
gpcgc: a green point cloud geometry coding method—0
GPLQ: A General, Practical, and Lightning QAT Method for Vision Transformers—0
Fast Large-Scale Discrete Optimization Based on Principal Coordinate Descent—0
GPTQT: Quantize Large Language Models Twice to Push the Efficiency—0
Fast Jet Tagging with MLP-Mixers on FPGAs—0
Fast Inference of Tree Ensembles on ARM Devices—0
Communication Efficient SGD via Gradient Sampling With Bayes Prior—0
GQ-Net: Training Quantization-Friendly Deep Networks—0
GQSA: Group Quantization and Sparsity for Accelerating Large Language Model Inference—0
GradFreeBits: Gradient Free Bit Allocation for Dynamic Low Precision Neural Networks—0
AddNet: Deep Neural Networks Using FPGA-Optimized Multipliers—0
WaveQ: Gradient-Based Deep Quantization of Neural Networks through Sinusoidal Adaptive Regularization—0
Gradient Based Method for the Fusion of Lattice Quantizers—0
Gradient-Based Post-Training Quantization: Challenging the Status Quo—0
Gradient Descent Quantizes ReLU Network Features—0
Fast Implementation of 4-bit Convolutional Neural Networks for Mobile Devices—0
Show:102550
← PrevPage 40 of 99Next →
All datasetsImageNetCIFAR-10Wiki-40BAgeDB-30CFP-FPCOCO (Common Objects in Context)IJB-BIJB-CKnowledge-based:LFW

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1FQ-ViT (ViT-L)Top-1 Accuracy (%)85.03—Unverified
2FQ-ViT (ViT-B)Top-1 Accuracy (%)83.31—Unverified
3FQ-ViT (Swin-B)Top-1 Accuracy (%)82.97—Unverified
4FQ-ViT (Swin-S)Top-1 Accuracy (%)82.71—Unverified
5FQ-ViT (DeiT-B)Top-1 Accuracy (%)81.2—Unverified
6FQ-ViT (Swin-T)Top-1 Accuracy (%)80.51—Unverified
7FQ-ViT (DeiT-S)Top-1 Accuracy (%)79.17—Unverified
8Xception W8A8Top-1 Accuracy (%)78.97—Unverified
9ADLIK-MO-ResNet50-W4A4Top-1 Accuracy (%)77.88—Unverified
10ADLIK-MO-ResNet50-W3A4Top-1 Accuracy (%)77.34—Unverified
#ModelMetricClaimedVerifiedStatus
13DCNN_VIVA_3MAP160,327.04—Unverified
2DTQMAP0.79—Unverified
#ModelMetricClaimedVerifiedStatus
1OutEffHop-Bert_basePerplexity6.3—Unverified
2OutEffHop-Bert_basePerplexity6.21—Unverified
#ModelMetricClaimedVerifiedStatus
1Accuracy98.13—Unverified
#ModelMetricClaimedVerifiedStatus
1Accuracy92.92—Unverified
#ModelMetricClaimedVerifiedStatus
1SSD ResNet50 V1 FPN 640x640MAP34.3—Unverified
#ModelMetricClaimedVerifiedStatus
1TAR @ FAR=1e-495.13—Unverified
#ModelMetricClaimedVerifiedStatus
1TAR @ FAR=1e-496.38—Unverified
#ModelMetricClaimedVerifiedStatus
13DCNN_VIVA_5All84,809,664—Unverified
#ModelMetricClaimedVerifiedStatus
1Accuracy99.8—Unverified