Test-Time Adaptation of Quantized ViTs via Single-Pass Quantizer-Aligned Recalibration
cs.CV, cs.AI
Submitted: 2026-10-06
Updated: 2026-10-06
Code: https://github.com/chahh9808/QuAR
Terminology
Sources
- Post-training 4-bit quantization of convolution networks for rapid-deployment
- Parameter-free Online Test-time Adaptation
- Test-Time Model Adaptation for Quantized Neural Networks
- ActMAD: Activation Matching to Align Distributions for Test-Time-Training
- NEO: No-Optimization Test-Time Adaptation through Latent Re-Centering
- FORGE: Forward-Only Test-Time Adaptation for Integer-Only Vision Models on Microcontrollers
- Improving robustness against common corruptions by covariate shift adaptation
- FOZO: Forward-Only Zeroth-Order Prompt Optimization for Test-Time Adaptation
- AdaLog: Post-Training Quantization for Vision Transformers with Adaptive Logarithm Quantizer
- PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models