arXiv:2605. 17232v5 Announce Type: replace Abstract: Discrete diffusion has become a leading framework for generative modeling in various applications including language, vision, and biology.
arXiv:2602. 15008v2 Announce Type: replace Abstract: Diffusion models over discrete spaces have recently shown striking empirical success, yet their theoretical foundations remain incomplete.
arXiv:2605. 17232v2 Announce Type: replace Abstract: Discrete diffusion has become a leading framework for generative modeling in various applications including language, vision, and biology.
arXiv:2610. 02663v1 Announce Type: cross Abstract: Despite the remarkable empirical success of flow-matching models, their statistical generalization guarantees remain underdeveloped.
arXiv:2607. 26285v1 Announce Type: cross Abstract: Two central challenges in diffusion-based sampling are the theoretical one of understanding their remarkable effectiveness even in high-dimensional settings, and the practical one of designing algorithms with certified performance guarantees.
The paper compares the parallelism capabilities of three diffusion large language model paradigms—masked, uniform, and Gaussian diffusion. It proves that uniform and Gaussian diffusion can sample with a number of forward passes scaling with the dual total correlation of the distribution, potentially much less than the context length, whereas masked diffusion may require more passes. The study establishes a provable separation in parallelism, showing that masked diffusion’s critical windows are asymptotically narrower than those of the other two approaches.
The paper presents new Kullback–Leibler convergence guarantees for score diffusion models that avoid restrictive Lipschitz assumptions on the score function. By analyzing models with fixed step size derived from the Ornstein–Uhlenbeck semigroup and its kinetic counterpart, the authors derive simple, sharp KL bounds that hold for any data distribution with finite Fisher information relative to the standard Gaussian. This extends theoretical guarantees to a broader class of data distributions without requiring early stopping or compact support.
Masked diffusion models (MDMs) generate discrete data by iteratively unmasking multiple tokens in parallel using a learned mask predictor. The paper provides end‑to‑end statistical guarantees for transformer‑based MDMs, showing that when the data distribution is well approximated by mixtures of hidden Markov trees, the model can achieve ε‑accurate sampling in KL divergence with ∫O(LK_par/ε) training samples, where K_par is polynomial in the sequence length, vocabulary size, number of hidden states, and mixture components. The transformer network and sampling procedure adapt automatically to the unknown latent structure without requiring prior knowledge of tree topologies or component parameters.
arXiv:2608. 13520v1 Announce Type: cross Abstract: We study masking diffusion for discrete sampling and introduce a path-resolved measure of data geometry called the \emph{unmasking growth complexity} ({\textsf{UGC}\xspace}).
arXiv:2607. 07967v1 Announce Type: cross Abstract: Diffusion-based policies have recently emerged as powerful policy parameterizations for reinforcement learning, representing state-conditioned action distributions as terminal laws of diffusion processes with parameterized drifts.
arXiv:2606. 07325v1 Announce Type: cross Abstract: We study the minimax rate of estimating a future value $\mu_{t_n+h}$ of a curve $t\mapsto\mu_t$ in the $2$-Wasserstein space $\mathcal{P}_2(\mathbb{R}^d)$ from finitely many noisy snapshots of its past, under an adiabatic bound $\|\nabla_t^k v\|\le\varepsilon$ on the $k$-th covariant derivative of the velocity field.
arXiv:2608. 06337v1 Announce Type: cross Abstract: A monotone adversary observes an i.