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 51–100 of 4925 papers

TitleStatusHype
LLM Inference Unveiled: Survey and Roofline Model InsightsCode4
BitDistiller: Unleashing the Potential of Sub-4-Bit LLMs via Self-DistillationCode4
QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice CodebooksCode4
Large Language Models for Time Series: A SurveyCode4
Fast Inference of Mixture-of-Experts Language Models with OffloadingCode4
Leveraging Speculative Sampling and KV-Cache Optimizations Together for Generative AI using OpenVINOCode4
Efficient Post-training Quantization with FP8 FormatsCode4
Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMsCode4
INT2.1: Towards Fine-Tunable Quantized Large Language Models with Error Correction through Low-Rank AdaptationCode4
SparseGPT: Massive Language Models Can Be Accurately Pruned in One-ShotCode4
The case for 4-bit precision: k-bit Inference Scaling LawsCode4
FP8 Formats for Deep LearningCode4
The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language ModelsCode4
Link and code: Fast indexing with graphs and compact regression codesCode4
Billion-scale similarity search with GPUsCode4
Polysemous codesCode4
Highly Compressed Tokenizer Can Generate Without TrainingCode3
TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and GenerationCode3
Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMsCode3
ParetoQ: Scaling Laws in Extremely Low-bit LLM QuantizationCode3
HAC++: Towards 100X Compression of 3D Gaussian SplattingCode3
MuQ: Self-Supervised Music Representation Learning with Mel Residual Vector QuantizationCode3
A Survey on Large Language Model Acceleration based on KV Cache ManagementCode3
A Survey on Inference Optimization Techniques for Mixture of Experts ModelsCode3
VidTok: A Versatile and Open-Source Video TokenizerCode3
APOLLO: SGD-like Memory, AdamW-level PerformanceCode3
TokenFlow: Unified Image Tokenizer for Multimodal Understanding and GenerationCode3
XQ-GAN: An Open-source Image Tokenization Framework for Autoregressive GenerationCode3
Scaling Transformers for Low-Bitrate High-Quality Speech CodingCode3
Pushing the Limits of Large Language Model Quantization via the Linearity TheoremCode3
Addressing Representation Collapse in Vector Quantized Models with One Linear LayerCode3
Data Generation for Hardware-Friendly Post-Training QuantizationCode3
COAT: Compressing Optimizer states and Activation for Memory-Efficient FP8 TrainingCode3
DPLM-2: A Multimodal Diffusion Protein Language ModelCode3
Latent Action Pretraining from VideosCode3
FlatQuant: Flatness Matters for LLM QuantizationCode3
ImageFolder: Autoregressive Image Generation with Folded TokensCode3
TCSinger: Zero-Shot Singing Voice Synthesis with Style Transfer and Multi-Level Style ControlCode3
BigCodec: Pushing the Limits of Low-Bitrate Neural Speech CodecCode3
TinyAgent: Function Calling at the EdgeCode3
Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language ModelCode3
ABQ-LLM: Arbitrary-Bit Quantized Inference Acceleration for Large Language ModelsCode3
Compact 3D Gaussian Splatting for Static and Dynamic Radiance FieldsCode3
Integer-Valued Training and Spike-Driven Inference Spiking Neural Network for High-performance and Energy-efficient Object DetectionCode3
Fast Matrix Multiplications for Lookup Table-Quantized LLMsCode3
EfficientQAT: Efficient Quantization-Aware Training for Large Language ModelsCode3
Image and Video Tokenization with Binary Spherical QuantizationCode3
CV-VAE: A Compatible Video VAE for Latent Generative Video ModelsCode3
Ditto: Quantization-aware Secure Inference of Transformers upon MPCCode3
MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of ExpertsCode3
Show:102550
← PrevPage 2 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