Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 18 of 18 for “"Discrete Diffusion"”.
-
Discrete Diffusion for Text Infilling
… in parallelizability and flexible control. Discrete diffusion models offer a promising alternative paradigm, leveraging iterative refinement and potentially enabling bidirectional context use, parallel generation, and flexible prompting. However, existing discrete text diffusion models …
-
Stability and inference in discrete diffusion scale-spaces
… by viewing it as a solution to a certain diffusion-inspired evolutionary partial differential equation (PDE). Such a model is known as the diffusion scale-space and it possesses two long-standing problems: (i) model analysis which aims at establishing stability and guarantees that …
-
EFFICIENT SIGNAL SYNTHESIS FOR DATA AUGMENTATION USING GENERATIVE AI
… with either a decoder-only transformer or a discrete diffusion model. The VQ-VAE encodes high-dimensional signals into discrete latent tokens, significantly reducing model complexity while enabling symbolic sequence modeling. These discrete representations are then modeled using …
-
Foundations of Diffusion Language Models
Diffusion models have recently emerged as a powerful alternative to autoregressive (AR) models for generative modeling, with strong results in continuous domains such as images and video. In the discrete setting of language, however, diffusion models still lag behind AR approaches in both …
-
From Score-Based Diffusion Models to Stochastic Flow Maps
This thesis studies diffusion models from the distributional and pathwise viewpoints. We first describe discrete diffusion models and their continuous-time formulation through stochastic differential equations. The reverse-time SDE depends on the score of the evolving density, which is estimated …
-
Knowledge-Centric Multimodal Intelligence: Understanding, Reasoning, and Generation
… intermediate outputs in both autoregressive and discrete-diffusion generation paradigms. For multimodal generation, I extend retrieval augmentation into image synthesis, incorporating fine-grained visual evidence that the text prompt alone cannot specify. Across visual question answering, …
-
Theory of charge and flux noise in superconducting wires
… impurities. This framework accounts for spin diffusion and spin-lattice relaxation, incorporating a discrete diffusion model to handle confinement effects and inhomogeneities. Analytical and numerical results show that the spin relaxation model aligns with experimental observations in aluminum …
-
Generative modeling and scalable test-time optimization using diffusion and flow models
… control, and structured anchoring. Modern diffusion-based generative models achieve remarkable quality, yet their use in inverse problems, personalization, editing, and structured generation remains limited by inefficient inference, heuristic guidance, and poor utilization of informative …
-
Multiscale modelling of continuum and discrete dynamics in materials with complicated microstructure
… a smooth macroscale model for a one-dimensional discrete diffusion system with rapidly varying microscale diffusivity, finite scale separation, and Dirichlet boundary conditions. I apply both centre manifold theory and homogenization theory to build the macroscale model. Both theories find same …
-
Constrained Diffusion: Applications to Image Generation, Manifold Learning, and Motion Planning
… foundations, extensions, and applications of diffusion modelling in generative tasks. Diffusion models have garnered significant attention due to their stability during training and superior performance compared to competing methods. In an attempt to make this work approachable for those not …
-
Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences
… they span graphs, 3D geometries, and discrete sequences; and practical systems must generalize beyond training distributions while remaining steerable at inference time toward domain-specific, often black-box, objectives. These demands call for robust AI that learns rich cross-modal …
-
Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences
… they span graphs, 3D geometries, and discrete sequences; and practical systems must generalize beyond training distributions while remaining steerable at inference time toward domain-specific, often black-box, objectives. These demands call for robust AI that learns rich cross-modal …
-
Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences
… they span graphs, 3D geometries, and discrete sequences; and practical systems must generalize beyond training distributions while remaining steerable at inference time toward domain-specific, often black-box, objectives. These demands call for robust AI that learns rich cross-modal …
-
Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences
… they span graphs, 3D geometries, and discrete sequences; and practical systems must generalize beyond training distributions while remaining steerable at inference time toward domain-specific, often black-box, objectives. These demands call for robust AI that learns rich cross-modal …
-
Guidance Conditions in Generative Modeling: Elevating Discriminative Capabilities and Controllability
Generative models, such as diffusion models and large language models (LLMs), play a crucial role in artificial intelligence due to their remarkable generative quality and versatility. These models leverage deep neural network architectures to sample new data points from complex data distributions …
-
Diffusion-controlled growth and transport in random composites
… thesis we consider three different models of diffusioncontrolled growth; phase-field models, diffusion-limited aggregation, and stochastic non-Laplacian growth models. Singular perturbation theory is employed to derive solutions to a phase-field model of solidification for a variety of …
-
Understanding theoretical and empirical foundations of diffusion models
… that have emerged over the past decade, diffusion models emerged as a powerful and principled framework, achieving state-of-the-art performance for continuous domains such as images and offering advantages like non-sequential reasoning and inference efficiency in discrete domains like …