2 citations · 8 across the 25 of their papers we have counts for
29 papers
Multimodal Co-Generation With Continuous Bitstream Diffusion
Georgios Batzolis, Georgios Kontos, Bruno Trentini +4
We show that continuous bitstream diffusion (CoBit), originally developed for language modeling, provides a unified framework for joint and conditional multimodal generation using…
Adaptive Second-Order Solvers for Fast Stochastic Diffusion Sampling
Ella Kemperman, Luca Ambrogioni
Diffusion models rely on numerical solvers requiring time-discretization, which has a large influence on the tradeoff between sampling cost and quality. However, the computational…
Domain shift-robust object detection with GenAI image editing
Isabel D. Stein, Thijs A. Eker, Sebastiaan P. Snel +4
Object detectors often degrade under domain shifts such as changes in lighting, weather, or occlusion. These shifts alter object appearance and expose a reliance on visual shortcut…
From Atoms to Entropy: Optimal Noise Allocation for Diffusion Training in the Convex Regime
Luca Ambrogioni, Giulio Franzese, Alberto Foresti +7
How should a diffusion model decide which noise levels to train on, and how much? Despite the importance of this choice, current noise schedules are based largely on heuristics or…
CoBit: Language Modeling with Bitstream Diffusion
Georgios Batzolis, Mark Girolami, Luca Ambrogioni
Diffusion language models (DLMs) promise parallel, order-agnostic generation, but on standard benchmarks they have historically lagged behind autoregressive models in sample qualit…
Entropy Across the Bridge: Conditional-Marginal Discretization for Flow and Schrödinger Samplers
Bruno Trentini, Dejan Stancevic, Michael M. Bronstein +2
For a fixed flow-based generative model under a small inference budget, sample quality can depend strongly on where the sampler spends its few function evaluations. Flow matching a…
How Out-of-Equilibrium Phase Transitions can Seed Pattern Formation in Trained Diffusion Models
Luca Ambrogioni
Diffusion models generate structure by progressively transforming noise into data, yet the mechanisms underlying this transition remain poorly understood. In this work, we show tha…
Language Diffusion Models are Associative Memories Capable of Retrieving Unseen Data
Bao Pham, Mohammed J. Zaki, Luca Ambrogioni +2
When do language diffusion models memorize their training data, and how to quantitatively assess their true generative regime? We address these questions by showing that Uniform-ba…
Categorical Flow Maps
Daan Roos, Oscar Davis, Floor Eijkelboom +5
We introduce Categorical Flow Maps, a flow-matching method for accelerated few-step generation of categorical data via self-distillation. Building on recent variational formulation…
Improved Sampling Schedules for Discrete Diffusion Models
Alberto Foresti, Mustapha Bounoua, Giulio Franzese +2
Discrete diffusion models have emerged as a powerful paradigm for generative modeling on sequence data; however, the information-theoretic principles governing their reverse proces…
Noise Scheduling as Information-Guided Allocation in Diffusion Training
Gabriel Raya, Bac Nguyen, Georgios Batzolis +6
We introduce InfoNoise, an online adaptive noise schedule for diffusion training that reallocates optimization effort toward noise levels where denoising is most informative. Toget…
The Entropic Signature of Class Speciation in Diffusion Models
Florian Handke, Dejan Stančević, Felix Koulischer +2
Diffusion models do not recover semantic structure uniformly over time. Instead, samples transition from semantic ambiguity to class commitment within a narrow regime. Recent theor…
Emergence of Distortions in High-Dimensional Guided Diffusion Models
Enrico Ventura, Beatrice Achilli, Luca Ambrogioni +1
Classifier-free guidance (CFG) is the de facto standard for conditional sampling in diffusion models, yet it often reduces sample diversity. Using tools from statistical physics, w…
The Information Dynamics of Generative Diffusion
Dejan Stancevic, Luca Ambrogioni
Generative diffusion models have emerged as a powerful class of models in machine learning, yet a unified theoretical understanding of their operation is still developing. This pap…
CoVAE: Consistency Training of Variational Autoencoders
Gianluigi Silvestri, Luca Ambrogioni
Current state-of-the-art generative approaches frequently rely on a two-stage training procedure, where an autoencoder (often a VAE) first performs dimensionality reduction, follow…
Measuring Semantic Information Production in Generative Diffusion Models
Florian Handke, Félix Koulischer, Gabriel Raya +1
It is well known that semantic and structural features of the generated images emerge at different times during the reverse dynamics of diffusion, a phenomenon that has been connec…
Feedback Guidance of Diffusion Models
Felix Koulischer, Florian Handke, Johannes Deleu +2
While Classifier-Free Guidance (CFG) has become standard for improving sample fidelity in conditional diffusion models, it can harm diversity and induce memorization by applying co…
Memorization to Generalization: Emergence of Diffusion Models from Associative Memory
Bao Pham, Gabriel Raya, Matteo Negri +3
Dense Associative Memories (DenseAMs) are generalizations of Hopfield networks, which have superior information storage capacity and can store training data points (memories) at lo…
Entropic Time Schedulers for Generative Diffusion Models
Dejan Stancevic, Florian Handke, Luca Ambrogioni
The practical performance of generative diffusion models depends on the appropriate choice of the noise scheduling function, which can also be equivalently expressed as a time repa…
The Capacity of Modern Hopfield Networks under the Data Manifold Hypothesis
Beatrice Achilli, Luca Ambrogioni, Carlo Lucibello +2
We generalize the computation of the capacity of exponential Hopfield model from Lucibello and Mézard (2024) to more generic pattern ensembles, including binary patterns and patter…
VCT: Training Consistency Models with Variational Noise Coupling
Gianluigi Silvestri, Luca Ambrogioni, Chieh-Hsin Lai +2
Consistency Training (CT) has recently emerged as a strong alternative to diffusion models for image generation. However, non-distillation CT often suffers from high variance and i…
Memorization and Generalization in Generative Diffusion under the Manifold Hypothesis
Beatrice Achilli, Luca Ambrogioni, Carlo Lucibello +2
We study the memorization and generalization capabilities of Diffusion Models (DMs) when data lies on a structured latent manifold. Specifically, we consider a set of data poin…
Dynamic Negative Guidance of Diffusion Models
Felix Koulischer, Johannes Deleu, Gabriel Raya +2
Negative Prompting (NP) is widely utilized in diffusion models, particularly in text-to-image applications, to prevent the generation of undesired features. In this paper, we show…
Losing dimensions: Geometric memorization in generative diffusion
Beatrice Achilli, Enrico Ventura, Gianluigi Silvestri +5
Diffusion models power leading generative AI, but when and how they memorize training data, especially on low-dimensional manifolds, remains unclear. We find memorization emerges g…