IRIS unifies self-play fine-tuning under an interpolative Rényi objective with adaptive alpha scheduling and reports better benchmark scores than baselines while surpassing full supervised fine-tuning with only 13% of the annotated data.
Springer, 2016
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
The KL generalization error in unsupervised learning decomposes exactly into model error, data bias, and variance for e-flat models, with closed-form results for ε-PCA on isotropic Gaussians showing optimal rank at the noise floor and a three-regime phase diagram.
Self-supervised encoders prefer isotropic Gaussian latent states because the Information Bottleneck, recast as rate-distortion over the predictive manifold, makes these states optimal for target-neutral representations.
citing papers explorer
-
IRIS: Interpolative R\'enyi Iterative Self-play for Large Language Model Fine-Tuning
IRIS unifies self-play fine-tuning under an interpolative Rényi objective with adaptive alpha scheduling and reports better benchmark scores than baselines while surpassing full supervised fine-tuning with only 13% of the annotated data.
-
Information-Geometric Decomposition of Generalization Error in Unsupervised Learning
The KL generalization error in unsupervised learning decomposes exactly into model error, data bias, and variance for e-flat models, with closed-form results for ε-PCA on isotropic Gaussians showing optimal rank at the noise floor and a three-regime phase diagram.
-
Why Self-Supervised Encoders Want to Be Normal
Self-supervised encoders prefer isotropic Gaussian latent states because the Information Bottleneck, recast as rate-distortion over the predictive manifold, makes these states optimal for target-neutral representations.