Imaginative Generative AI: Crossing the Entropy Wall into Worlds Beyond Imitation
2026-08-10 • Machine Learning
Machine LearningArtificial IntelligenceComputer Vision and Pattern Recognition
AI summaryⓘ
The authors introduce Imaginative Generative AI (IGA), a method that aims to control how diverse generated outputs are, either matching or going beyond the diversity found in the training data. They measure diversity using a mathematical concept called spectral entropy, which looks at how spread out the generated data is in a feature space. IGA can fix situations where the generator produces less variety than the original data (called diversity repair) or deliberately create more diverse outputs than the data allows (called imaginative generation). Their approach works with existing generative models without retraining and was tested successfully on both synthetic and real images.
Generative AIDiversity RepairSpectral EntropyKernel Covariance OperatorRepresentation SpaceEntropy WallExponential TiltScore-based ModelsDiffusion ModelsKL Divergence
Authors
Hossein Goli, Farzan Farnia, Amin Gohari
Abstract
Generative AI models are primarily designed to imitate the data distribution, an objective that neither corrects diversity lost by a learned generator nor defines how generation should extend beyond the diversity of the data itself. We introduce Imaginative Generative AI (IGA), a framework that makes diversity part of the target-distribution design problem: among distributions close to a reference, IGA selects one whose spectral diversity reaches a prescribed level. Diversity is measured by the von Neumann entropy of the generated distribution's kernel covariance operator in a fixed representation space, providing a reference-free representation-guided measure of how broadly probability mass occupies embedding directions. The spectral entropy of the population data distribution defines an Entropy Wall. Below the wall, IGA performs diversity repair, recovering variation that a learned generator has lost while remaining within the diversity level of the data. Beyond the wall, the data distribution itself becomes infeasible, and IGA deliberately departs from it to produce distributions with greater representation-relative spectral diversity, an operational notion of imaginative generation. These regimes form a single regularization path from imitation to imagination and define an i.i.d. target distribution at each prescribed diversity level. We develop the theory of this entropy-constrained projection and show that, under a KL anchor to a pretrained generator, the optimum satisfies a self-consistent exponential-tilt relation. This characterization leads to IGA Guidance, a retraining-free inference-time method for score-based and diffusion models, including DDPM and DDIM samplers. Experiments on synthetic and vision benchmarks demonstrate diversity repair below the Entropy Wall and controlled spectral extrapolation beyond it.