paper-conference

Latent diffusion models offer an attractive alternative to discrete diffusion for non-autoregressive text generation by operating on …

Midpoint Generative Models learn one-step generators from scratch using a divergence built on the symmetry of flow matching and randomly flipped interpolations.

Event sequences are chains of random events, each described by heterogeneous (numerical and categorical) features. The problem of event …

TUBE provides an upper bound on log-likelihood for discrete diffusion language models. Together with the ELBO, it brackets the true value and enables principled comparisons with autoregressive models.

We reveal several limitations of existing solvers for the continuous Gromov-Wasserstein problem and pave the way to get rid of them by proposing a novel solver.

The work unifies significant parts of the generative modeling landscape. It shows that CFM and (forward-only) IFM are dual perspectives of the same approach, while positioning general IFM as a broader, more expressive paradigm that encompasses and extends beyond CFM. This bridge deepens theoretical understanding and opens avenues for practical innovation in both lines of research.

This paper introduces IDLM, a method that accelerates discrete Diffusion Language Models by adapting inverse distillation techniques, achieving 4x-64x faster inference while maintaining generative quality through theoretically-validated unique solutions and stable gradient relaxations.

We propose a novel light learning algorithm for semi-supervised domain translation which can utilize both paired and unpaired data.

We introduce a novel distillation technique for ResShift, one of the best diffusion models for real-world image super-resolution.

This paper introduces Inverse Poisson Flow Matching (IPFM), a new distillation method that significantly accelerates electrostatic generative models like PFGM by learning a generator to match the teacher’s electrostatic field, enabling high-quality sample generation in few steps.