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 101–150 of 4925 papers

TitleStatusHype
PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language ModelsCode3
The Unreasonable Ineffectiveness of the Deeper LayersCode3
HAC: Hash-grid Assisted Context for 3D Gaussian Splatting CompressionCode3
GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian SplattingCode3
Behavior Generation with Latent ActionsCode3
NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion ModelsCode3
IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens IntactCode3
Language-Codec: Bridging Discrete Codec Representations and Speech Language ModelsCode3
OneBit: Towards Extremely Low-bit Large Language ModelsCode3
BiLLM: Pushing the Limit of Post-Training Quantization for LLMsCode3
KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV CacheCode3
KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache QuantizationCode3
FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algorithm-System Co-DesignCode3
Inferflow: an Efficient and Highly Configurable Inference Engine for Large Language ModelsCode3
RoSA: Accurate Parameter-Efficient Fine-Tuning via Robust AdaptationCode3
TinyGPT-V: Efficient Multimodal Large Language Model via Small BackbonesCode3
Compact 3D Scene Representation via Self-Organizing Gaussian GridsCode3
MotionGPT: Human Motion as a Foreign LanguageCode3
High-Fidelity Audio Compression with Improved RVQGANCode3
LLM-QAT: Data-Free Quantization Aware Training for Large Language ModelsCode3
Autoregressive Image Generation using Residual QuantizationCode3
8-bit Optimizers via Block-wise QuantizationCode3
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech RepresentationsCode3
MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group QuantizationCode2
MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group QuantizationCode2
any4: Learned 4-bit Numeric Representation for LLMsCode2
AnalogNAS-Bench: A NAS Benchmark for Analog In-Memory ComputingCode2
From Tiny Machine Learning to Tiny Deep Learning: A SurveyCode2
DETRPose: Real-time end-to-end transformer model for multi-person pose estimationCode2
BitVLA: 1-bit Vision-Language-Action Models for Robotics ManipulationCode2
RecGPT: A Foundation Model for Sequential RecommendationCode2
Model-Preserving Adaptive RoundingCode2
FlowSE: Efficient and High-Quality Speech Enhancement via Flow MatchingCode2
Efficient Speech Language Modeling via Energy Distance in Continuous Latent SpaceCode2
GuidedQuant: Large Language Model Quantization via Exploiting End Loss GuidanceCode2
Diffusion Model Quantization: A ReviewCode2
An Empirical Study of Qwen3 QuantizationCode2
Softpick: No Attention Sink, No Massive Activations with Rectified SoftmaxCode2
Enhancing Autonomous Driving Systems with On-Board Deployed Large Language ModelsCode2
Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning ModelsCode2
GPTAQ: Efficient Finetuning-Free Quantization for Asymmetric CalibrationCode2
Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space ModelsCode2
Harmonizing Visual Representations for Unified Multimodal Understanding and GenerationCode2
GENIUS: A Generative Framework for Universal Multimodal SearchCode2
BitDecoding: Unlocking Tensor Cores for Long-Context LLMs Decoding with Low-Bit KV CacheCode2
D2GV: Deformable 2D Gaussian Splatting for Video Representation in 400FPSCode2
BHViT: Binarized Hybrid Vision TransformerCode2
Train Small, Infer Large: Memory-Efficient LoRA Training for Large Language ModelsCode2
QuEST: Stable Training of LLMs with 1-Bit Weights and ActivationsCode2
Massive Values in Self-Attention Modules are the Key to Contextual Knowledge UnderstandingCode2
Show:102550
← PrevPage 3 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