6 papers
NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment
Jisung Hwang, Yunhong Min, Jaihoon Kim +2
We introduce the Noise-Tilted Reverse Kernel (NTRK), a reward-guided diffusion sampler that injects reward gradients through the noise term, leaving the pretrained reverse kernel u…
Gradient Preconditioning for Efficient and Reliable Reward-Guided Generation
Jisung Hwang, Minhyuk Sung
We propose a gradient preconditioning method that makes reward-guided generation with one-step generative models both efficient and reliable. Test-time noise optimization can unloc…
PairFlow: Closed-Form Source-Target Coupling for Few-Step Generation in Discrete Flow Models
Mingue Park, Jisung Hwang, Seungwoo Yoo +2
We introduce , a lightweight preprocessing step for training Discrete Flow Models (DFMs) to achieve few-step sampling without requiring a pretrained teacher. DFM…
Neural Green's Functions
Seungwoo Yoo, Kyeongmin Yeo, Jisung Hwang +1
We introduce Neural Green's Function, a neural solution operator for linear partial differential equations (PDEs) whose differential operators admit eigendecompositions. Inspired b…
Inference-Time Scaling for Flow Models via Stochastic Generation and Rollover Budget Forcing
Jaihoon Kim, Taehoon Yoon, Jisung Hwang +1
We propose an inference-time scaling approach for pretrained flow models. Recently, inference-time scaling has gained significant attention in LLMs and diffusion models, improving…
Moment- and Power-Spectrum-Based Gaussianity Regularization for Text-to-Image Models
Jisung Hwang, Jaihoon Kim, Minhyuk Sung
We propose a novel regularization loss that enforces standard Gaussianity, encouraging samples to align with a standard Gaussian distribution. This facilitates a range of downstrea…