AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Is Bin Generation Indispensable? A Bin-Generation-Free Dataset Quantization via Semantic Perspective

Maijie Deng, Yuhua Li, Yixiong Zou, Yao Wu, Chenru Ma

CVPR 2026 · Architectures and efficiency › Quantization · approach: Post-training weight quantization

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace bin generation with KNN neighbor identificationBGFDQ (w/o SPD) · CIFAR-10 runtime (rho=20%)total runtime (s): 1248 -> 1052Improved
add semantic-shift patch dropping (SPD) (ablation)BGFDQ · CIFAR-100 (rho=20%)validation accuracy: 62.8 -> 66.0Improved
add semantic-shift patch dropping (SPD) (ablation)BGFDQ · CIFAR-10 runtime (rho=20%)total runtime (s): 1052 -> 2465Hurt
use neighbor-aware coreset selection instead of bin generation + random selection (ablation)coreset selection · CIFAR-10, 5K samplesvalidation accuracy: 75.9 -> 76.4Improved
add bin generation before random selection (ablation)random selection · CIFAR-10 and CIFAR-100validation accuracy: 0.1% average improvementNo clear effect
raise vanilla patch drop ratio from 25% to 40% (ablation)patch dropping · CIFAR-10 random subsetsaccuracy change vs unquantizedHurt
use adaptive SPD (avg 40% drop) instead of vanilla 40% patch dropping (ablation)patch dropping · CIFAR-10 random subsetsaccuracy change vs unquantizedImproved
use BGFDQ instead of DQ coresets for ViT trainingcoreset · CIFAR-100 ViT (rho=20%)validation accuracy: 30.8 -> 35.7Improved
use BGFDQ on Fakeddit where bin-generation methods run out of memorycoreset selection · Fakeddit (rho=1%)validation accuracy: 83.3 -> 83.8Improved
vary neighbor count K (ablation)neighbor identification · CIFAR-10validation accuracyNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy