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 251–300 of 4925 papers

TitleStatusHype
MAUVE Scores for Generative Models: Theory and PracticeCode2
Compressing Volumetric Radiance Fields to 1 MBCode2
A Closer Look at Hardware-Friendly Weight QuantizationCode2
I-ViT: Integer-only Quantization for Efficient Vision Transformer InferenceCode2
On-Device Training Under 256KB MemoryCode2
ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale TransformersCode2
Re-parameterizing Your Optimizers rather than ArchitecturesCode2
VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel DecoderCode2
BMInf: An Efficient Toolkit for Big Model Inference and TuningCode2
4-bit Conformer with Native Quantization Aware Training for Speech RecognitionCode2
QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training QuantizationCode2
QuantumNAT: Quantum Noise-Aware Training with Noise Injection, Quantization and NormalizationCode2
hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning DevicesCode2
FBGEMM: Enabling High-Performance Low-Precision Deep Learning InferenceCode2
Fast convolutional neural networks on FPGAs with hls4mlCode2
I-BERT: Integer-only BERT QuantizationCode2
Binary Neural Networks: A SurveyCode2
Neural Network Compression Framework for fast model inferenceCode2
HAQ: Hardware-Aware Automated Quantization with Mixed PrecisionCode2
Bolt: Accelerated Data Mining with Fast Vector CompressionCode2
Compress Any Segment Anything Model (SAM)Code1
CycleVAR: Repurposing Autoregressive Model for Unsupervised One-Step Image TranslationCode1
Q-resafe: Assessing Safety Risks and Quantization-aware Safety Patching for Quantized Large Language ModelsCode1
CommVQ: Commutative Vector Quantization for KV Cache CompressionCode1
FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix ApproximationCode1
Speculative Decoding Meets Quantization: Compatibility Evaluation and Hierarchical Framework DesignCode1
TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache OptimizationCode1
Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM CompressionCode1
FP4 All the Way: Fully Quantized Training of LLMsCode1
Mind the Gap: A Practical Attack on GGUF QuantizationCode1
PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMsCode1
DVD-Quant: Data-free Video Diffusion Transformers QuantizationCode1
UniTTS: An end-to-end TTS system without decoupling of acoustic and semantic informationCode1
Optimizing Binary and Ternary Neural Network Inference on RRAM Crossbars using CIM-ExplorerCode1
Quaff: Quantized Parameter-Efficient Fine-Tuning under Outlier Spatial Stability HypothesisCode1
Fine-tuning Quantized Neural Networks with Zeroth-order OptimizationCode1
GenoArmory: A Unified Evaluation Framework for Adversarial Attacks on Genomic Foundation ModelsCode1
Accurate KV Cache Quantization with Outlier Tokens TracingCode1
EA-3DGS: Efficient and Adaptive 3D Gaussians with Highly Enhanced Quality for outdoor scenesCode1
Analog Foundation ModelsCode1
Continuous Visual Autoregressive Generation via Score MaximizationCode1
MxMoE: Mixed-precision Quantization for MoE with Accuracy and Performance Co-DesignCode1
RGB-Event Fusion with Self-Attention for Collision PredictionCode1
Fast and Low-Cost Genomic Foundation Models via Outlier RemovalCode1
NoWag: A Unified Framework for Shape Preserving Compression of Large Language ModelsCode1
Hierarchical Vector Quantized Graph Autoencoder with Annealing-Based Code SelectionCode1
Task-Circuit Quantization: Leveraging Knowledge Localization and Interpretability for CompressionCode1
APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision TransformersCode1
MiLo: Efficient Quantized MoE Inference with Mixture of Low-Rank CompensatorsCode1
MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and QuantizationCode1
Show:102550
← PrevPage 6 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