Using two class labels in text prompts to generate synthetic calibration images improves data-free post-training quantization accuracy, especially in low-bit settings.
On generaliza- tion error bounds of noisy gradient methods for non-convex learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Enhancing Generalization in Data-free Quantization via Mixup-class Prompting
Using two class labels in text prompts to generate synthetic calibration images improves data-free post-training quantization accuracy, especially in low-bit settings.