DiffICL breaks the quality-privacy tradeoff in small-data tabular synthesis by using in-context learning on pretrained structural priors to generate data that is both higher quality and less memorizing of training samples.
arXiv preprint arXiv:1806.03384 (2018)
4 Pith papers cite this work. Polarity classification is still indexing.
fields
cs.LG 4years
2026 4verdicts
UNVERDICTED 4representative citing papers
Hyperparameter-optimized generative models augment scarce flight diversion records and substantially improve prediction accuracy over real data alone.
Fully connected neural network with randomized loss synthesizes real-world tabular data distributions from Gaussian noise faster than state-of-the-art deep generative models.
Synthetic flight data generated by TVAE and Gaussian Copula models supports flight delay prediction models with accuracy comparable to real data.
citing papers explorer
-
Breaking the Quality-Privacy Tradeoff in Tabular Data Generation via In-Context Learning
DiffICL breaks the quality-privacy tradeoff in small-data tabular synthesis by using in-context learning on pretrained structural priors to generate data that is both higher quality and less memorizing of training samples.
-
Generative Augmentation of Imbalanced Flight Records for Flight Diversion Prediction: A Multi-objective Optimisation Framework
Hyperparameter-optimized generative models augment scarce flight diversion records and substantially improve prediction accuracy over real data alone.
-
Synthesizing real-world distributions from high-dimensional Gaussian Noise with Fully Connected Neural Network
Fully connected neural network with randomized loss synthesizes real-world tabular data distributions from Gaussian noise faster than state-of-the-art deep generative models.
-
Synthetic Flight Data Generation Using Generative Models
Synthetic flight data generated by TVAE and Gaussian Copula models supports flight delay prediction models with accuracy comparable to real data.