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 2551–2600 of 4925 papers

TitleStatusHype
Improving Multi-generation Robustness of Learned Image Compression—0
A Faster Approach to Spiking Deep Convolutional Neural Networks—0
L-GreCo: Layerwise-Adaptive Gradient Compression for Efficient and Accurate Deep LearningCode1
Empirical Evaluation of Post-Training Quantization Methods for Language Tasks—0
Machine Unlearning of Federated ClustersCode1
Multiresolution Signal Processing of Financial Market Objects—0
Fast DistilBERT on CPUs—0
Collaborative Multi-Teacher Knowledge Distillation for Learning Low Bit-width Deep Neural Networks—0
Neural Networks with Quantization ConstraintsCode0
Too Brittle To Touch: Comparing the Stability of Quantization and Distillation Towards Developing Lightweight Low-Resource MT ModelsCode1
Constrained Approximate Similarity Search on Proximity Graph—0
Zero-Shot Learning of a Conditional Generative Adversarial Network for Data-Free Network Quantization—0
Broken Neural Scaling LawsCode1
Multi-rate adaptive transform coding for video compression—0
Streaming Parrotron for on-device speech-to-speech conversion—0
Legal-Tech Open Diaries: Lesson learned on how to develop and deploy light-weight models in the era of humongous Language Models—0
Langevin dynamics based algorithm e-THO POULA for stochastic optimization problems with discontinuous stochastic gradientCode0
BASQ: Branch-wise Activation-clipping Search Quantization for Sub-4-bit Neural NetworksCode0
Mixed Precision Quantization to Tackle Gradient Leakage Attacks in Federated Learning—0
PoseGPT: Quantization-based 3D Human Motion Generation and ForecastingCode1
Discrete Cross-Modal Alignment Enables Zero-Shot Speech TranslationCode0
ODG-Q: Robust Quantization via Online Domain Generalization—0
Sub-8-bit quantization for on-device speech recognition: a regularization-free approach—0
HQNAS: Auto CNN deployment framework for joint quantization and architecture search—0
FIT: A Metric for Model Sensitivity—0
FAQS: Communication-efficient Federate DNN Architecture and Quantization Co-Search for personalized Hardware-aware Preferences—0
CAP: Correlation-Aware Pruning for Highly-Accurate Sparse Vision Models—0
Just Round: Quantized Observation Spaces Enable Memory Efficient Learning of Dynamic LocomotionCode0
Accelerating RNN-based Speech Enhancement on a Multi-Core MCU with Mixed FP16-INT8 Post-Training Quantization—0
Post-Training Quantization for Energy Efficient Realization of Deep Neural Networks—0
ScionFL: Efficient and Robust Secure Quantized Aggregation—0
Q-ViT: Accurate and Fully Quantized Low-bit Vision TransformerCode1
SQuAT: Sharpness- and Quantization-Aware Training for BERT—0
SNN-SC: A Spiking Semantic Communication Framework for Collaborative IntelligenceCode0
Statistical Modeling of Soft Error Influence on Neural Networks—0
Clustering the Sketch: A Novel Approach to Embedding Table CompressionCode1
Deep learning model compression using network sensitivity and gradients—0
Hybrid Inverted Index Is a Robust Accelerator for Dense RetrievalCode1
Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Images—0
AlphaTuning: Quantization-Aware Parameter-Efficient Adaptation of Large-Scale Pre-Trained Language Models—0
Learning to embed semantic similarity for joint image-text retrieval—0
A Closer Look at Hardware-Friendly Weight QuantizationCode2
Set2Box: Similarity Preserving Representation Learning of SetsCode1
Self-triggered Consensus of Multi-agent Systems with Quantized Relative State Measurements—0
IR2Net: Information Restriction and Information Recovery for Accurate Binary Neural NetworksCode0
Active Image IndexingCode1
GLM-130B: An Open Bilingual Pre-trained ModelCode6
Sampling Streaming Data with Parallel Vector Quantization -- PVQ—0
A Novel Light Field Coding Scheme Based on Deep Belief Network & Weighted Binary Images for Additive Layered Displays—0
Convolutional Neural Networks Quantization with Attention—0
Show:102550
← PrevPage 52 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