6 papers
D2PO: Optimizing Diffusion Samplers via Dynamic Preference
Jinkyu Kim, Jinyoung Choi, Bohyung Han
We propose D2PO (Dynamic Direct Preference Optimization), a principled framework for optimizing diffusion sampling policies with respect to timestep schedules and classifier-free g…
Probabilistic Signature Inversion: Learning Conditional Distributions from Truncated Signatures
Junoh Kang, Kiseop Lee, Bohyung Han
The signature transform is a principled feature map for continuous-time paths, valued for its uniqueness and universality. Recovering a path from its truncated signature is, howeve…
Score-Repellent Monte Carlo: Toward Efficient Non-Markovian Sampler with Constant Memory in General State Spaces
Jie Hu, Lingyun Chen, Geeho Kim +3
History-dependent sampling can reduce long-run Monte Carlo variance by discouraging redundant revisits, but existing schemes typically encode history through empirical measure on f…
ICM-SR: Image-Conditioned Manifold Regularization for Image Super-Resolution
Junoh Kang, Donghun Ryou, Bohyung Han
Real world image super-resolution (Real-ISR) often leverages the powerful generative priors of text-to-image diffusion models by regularizing the output to lie on their learned man…
STR-Match: Matching SpatioTemporal Relevance Score for Training-Free Video Editing
Junsung Lee, Junoh Kang, Bohyung Han
Previous text-guided video editing methods often suffer from temporal inconsistency, motion distortion, and-most notably-limited domain transformation. We attribute these limitatio…
Enhanced Diffusion Sampling via Extrapolation with Multiple ODE Solutions
Jinyoung Choi, Junoh Kang, Bohyung Han
Diffusion probabilistic models (DPMs), while effective in generating high-quality samples, often suffer from high computational costs due to their iterative sampling process. To ad…