6 papers
One-Sided Quantile Coupling for Flow Matching
Jin-Young Kim, So-Yoon Cho, Hyun-Gyoon Kim
Flow Matching trains continuous-time generative models by regressing the velocity field of a probability path between a simple source distribution and a target data distribution. T…
Understanding, Accelerating, and Improving MeanFlow Training
Jin-Young Kim, Hyojun Go, Lea Bogensperger +5
MeanFlow promises high-quality generative modeling in few steps, by jointly learning instantaneous and average velocity fields. Yet, the underlying training dynamics remain unclear…
SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis
Hyojun Go, Byeongjun Park, Jiho Jang +3
Text-based generation and editing of 3D scenes hold significant potential for streamlining content creation through intuitive user interactions. While recent advances leverage 3D G…
Toward Stable World Models: Measuring and Addressing World Instability in Generative Environments
Soonwoo Kwon, Jin-Young Kim, Hyojun Go +1
We present a novel study on enhancing the capability of preserving the content in world models, focusing on a property we term World Stability. Recent diffusion-based generative mo…
Denoising Task Difficulty-based Curriculum for Training Diffusion Models
Jin-Young Kim, Hyojun Go, Soonwoo Kwon +1
Diffusion-based generative models have emerged as powerful tools in the realm of generative modeling. Despite extensive research on denoising across various timesteps and noise lev…
Diffusion Model Patching via Mixture-of-Prompts
Seokil Ham, Sangmin Woo, Jin-Young Kim +3
We present Diffusion Model Patching (DMP), a simple method to boost the performance of pre-trained diffusion models that have already reached convergence, with a negligible increas…