SOTAVerified

Data Free Quantization

Data Free Quantization is a technique to achieve a highly accurate quantized model without accessing any training data.

Source: Qimera: Data-free Quantization with Synthetic Boundary Supporting Samples

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 11–20 of 37 papers

TitleStatusHype
Robustness-Guided Image Synthesis for Data-Free Quantization—0
Causal-DFQ: Causality Guided Data-free Network QuantizationCode0
Data-Free Quantization via Mixed-Precision Compensation without Fine-Tuning—0
QNNRepair: Quantized Neural Network Repair—0
Towards Accurate Post-training Quantization for Diffusion ModelsCode1
LLM-QAT: Data-Free Quantization Aware Training for Large Language ModelsCode3
Adaptive Data-Free QuantizationCode1
Rethinking Data-Free Quantization as a Zero-Sum GameCode0
ACQ: Improving Generative Data-free Quantization Via Attention Correction—0
Genie: Show Me the Data for QuantizationCode0
Show:102550
← PrevPage 2 of 4Next →

No leaderboard results yet.