7 papers
Improving score-based sampling via affine post-processing
Curtis McDonald, Kevin Rojas, Molei Tao +1
Sampling from a probability distribution with access only to its log-density is a foundational problem in statistics, machine learning, and the physical sciences. This problem is s…
Improving Reasoning for Diffusion Language Models via Group Diffusion Policy Optimization
Kevin Rojas, Jiahe Lin, Kashif Rasul +4
Diffusion language models (DLMs) enable parallel, order-agnostic generation with iterative refinement, offering a flexible alternative to autoregressive large language models (LLMs…
Improving Classifier-Free Guidance in Masked Diffusion: Low-Dim Theoretical Insights with High-Dim Impact
Kevin Rojas, Ye He, Chieh-Hsin Lai +3
Classifier-Free Guidance (CFG) is a widely used technique for conditional generation and improving sample quality in continuous diffusion models, and its extensions to discrete dif…
What Exactly Does Guidance Do in Masked Discrete Diffusion Models
He Ye, Rojas Kevin, Tao Molei
We study masked discrete diffusion models with classifier-free guidance (CFG). Assuming no score error nor discretization error, we derive an explicit solution to the guided revers…
Diffuse Everything: Multimodal Diffusion Models on Arbitrary State Spaces
Kevin Rojas, Yuchen Zhu, Sichen Zhu +2
Diffusion models have demonstrated remarkable performance in generating unimodal data across various tasks, including image, video, and text generation. On the contrary, the joint…
Zeroth-Order Sampling Methods for Non-Log-Concave Distributions: Alleviating Metastability by Denoising Diffusion
Ye He, Kevin Rojas, Molei Tao
This paper considers the problem of sampling from non-logconcave distribution, based on queries of its unnormalized density. It first describes a framework, Denoising Diffusion Mon…
DFU: scale-robust diffusion model for zero-shot super-resolution image generation
Alex Havrilla, Kevin Rojas, Wenjing Liao +1
Diffusion generative models have achieved remarkable success in generating images with a fixed resolution. However, existing models have limited ability to generalize to different…