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 1151–1200 of 4925 papers

TitleStatusHype
Leveraging Continuously Differentiable Activation Functions for Learning in Quantized Noisy EnvironmentsCode0
Learning Frequency-Specific Quantization Scaling in VVC for Standard-Compliant Task-driven Image CodingCode0
Learning Convolutional Transforms for Lossy Point Cloud Geometry CompressionCode0
Learning Compression from Limited Unlabeled DataCode0
Learning Bag-of-Features Pooling for Deep Convolutional Neural NetworksCode0
Learning Accurate Low-Bit Deep Neural Networks with Stochastic QuantizationCode0
Learning Accurate Performance Predictors for Ultrafast Automated Model CompressionCode0
Learning compact binary descriptors with unsupervised deep neural networksCode0
Adaptive Computation Modules: Granular Conditional Computation For Efficient InferenceCode0
An Integrated Approach to Produce Robust Models with High EfficiencyCode0
An Information-Theoretic Analysis of Self-supervised Discrete Representations of SpeechCode0
Boosting CNN-based primary quantization matrix estimation of double JPEG images via a classification-like architectureCode0
Language Models as Zero-shot Lossless Gradient Compressors: Towards General Neural Parameter Prior ModelsCode0
Learned transform compression with optimized entropy encodingCode0
Learning Physical-Layer Communication with Quantized FeedbackCode0
Linearly Converging Error Compensated SGDCode0
KP2Dtiny: Quantized Neural Keypoint Detection and Description on the EdgeCode0
KVTuner: Sensitivity-Aware Layer-wise Mixed Precision KV Cache Quantization for Efficient and Nearly Lossless LLM InferenceCode0
JPEG Inspired Deep LearningCode0
BlockDialect: Block-wise Fine-grained Mixed Format Quantization for Energy-Efficient LLM InferenceCode0
Joint Pruning and Channel-wise Mixed-Precision Quantization for Efficient Deep Neural NetworksCode0
An exploration of the effect of quantisation on energy consumption and inference time of StarCoder2Code0
Joint Maximum Purity Forest with Application to Image Super-ResolutionCode0
Just Round: Quantized Observation Spaces Enable Memory Efficient Learning of Dynamic LocomotionCode0
Langevin dynamics based algorithm e-THO POULA for stochastic optimization problems with discontinuous stochastic gradientCode0
Iterative Training: Finding Binary Weight Deep Neural Networks with Layer BinarizationCode0
Is PGD-Adversarial Training Necessary? Alternative Training via a Soft-Quantization Network with Noisy-Natural Samples OnlyCode0
IR2Net: Information Restriction and Information Recovery for Accurate Binary Neural NetworksCode0
Forward and Backward Information Retention for Accurate Binary Neural NetworksCode0
I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs QuantizationCode0
BitMoD: Bit-serial Mixture-of-Datatype LLM AccelerationCode0
Integral Human Pose RegressionCode0
Integer Quantization for Deep Learning Inference: Principles and Empirical EvaluationCode0
Integrated Encoding and Quantization to Enhance Quanvolutional Neural NetworksCode0
Investigating the Impact of Quantization Methods on the Safety and Reliability of Large Language ModelsCode0
FTBNN: Rethinking Non-linearity for 1-bit CNNs and Going BeyondCode0
Instance-Aware Dynamic Neural Network QuantizationCode0
Improving Self-Supervised Learning-based MOS Prediction NetworksCode0
In-Context Learning for MIMO Equalization Using Transformer-Based Sequence ModelsCode0
Incremental Network Quantization: Towards Lossless CNNs with Low-Precision WeightsCode0
Improving Neural Network Quantization without Retraining using Outlier Channel SplittingCode0
AdaBits: Neural Network Quantization with Adaptive Bit-WidthsCode0
BinaryRelax: A Relaxation Approach For Training Deep Neural Networks With Quantized WeightsCode0
An efficient and straightforward online quantization method for a data stream through remove-birth updatingCode0
AdaBin: Improving Binary Neural Networks with Adaptive Binary SetsCode0
An Edge Computing-Based Solution for Real-Time Leaf Disease Classification using Thermal ImagingCode0
Improved Gradient based Adversarial Attacks for Quantized NetworksCode0
Improved Knowledge Distillation for Crowd Counting on IoT DeviceCode0
Improving Robustness Against Stealthy Weight Bit-Flip Attacks by Output Code MatchingCode0
Image Hashing by Minimizing Discrete Component-wise Wasserstein DistanceCode0
Show:102550
← PrevPage 24 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