A gated hierarchical image-enhancement network trained with quantization-aware training maintains high visual fidelity after low-precision conversion while keeping low computational cost on mobile devices.
Improving generalization in visual reasoning via self-ensemble
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A quantized INT8 SISR model with extract-refine-upsample design and teacher-guided three-stage training achieves 29.79 dB PSNR and 0.8634 SSIM on the MAI 2026 Quantized 4K challenge under mobile INT8 constraints.
citing papers explorer
-
Bridging the Training-Deployment Gap: Gated Encoding and Multi-Scale Refinement for Efficient Quantization-Aware Image Enhancement
A gated hierarchical image-enhancement network trained with quantization-aware training maintains high visual fidelity after low-precision conversion while keeping low computational cost on mobile devices.
-
Efficient INT8 Single-Image Super-Resolution via Deployment-Aware Quantization and Teacher-Guided Training
A quantized INT8 SISR model with extract-refine-upsample design and teacher-guided three-stage training achieves 29.79 dB PSNR and 0.8634 SSIM on the MAI 2026 Quantized 4K challenge under mobile INT8 constraints.