Rethinking Refinement: Correcting Generative Bias without Noise Injection
Generative models, including diffusion and flow-based models, often exhibit systematic biases that degrade sample quality, particularly in high-dimensional settings. We revisit refinement methods and show that effective bias correction can be achieved as a post-hoc procedure, without noise injection or multi-step resampling of the sampling process. We propose a flow-matching-based \textbf{Bi-stage Flow Refinement (BFR)} framework with two refinement strategies operating at different stages: latent space alignment for approximately invertible generators and data space refinement trained with lightweight augmentations. Unlike previous refiners that perturb sampling dynamics, BFR preserves the original ODE trajectory and applies deterministic corrections to generated samples. Experiments on MNIST, CIFAR-10, and FFHQ at 256x256 resolution demonstrate consistent improvements in fidelity and coverage; notably, starting from base samples with FID 3.95, latent space refinement achieves a \textbf{state-of-the-art} FID of \textbf{1.46} on MNIST using only a single additional function evaluation (1-NFE), while maintaining sample diversity.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
qecGPT: decoding Quantum Error-correcting Codes with Generative Pre-trained Transformers
We propose a general framework for decoding quantum error-correcting codes with generative modeling. The model utilizes autoregressive neural networks, specifically Transformers, to learn the joint probability of logical…
2kSC2GAN: Rethinking Entanglement by Self-correcting Correlated GAN Space
Generative Adversarial Networks (GANs) can synthesize realistic images, with the learned latent space shown to encode rich semantic information with various interpretable directions. However, due to the unstructured natu…
AttributeDisentanglementRePL: Pseudo-label Refinement for Semi-supervised LiDAR Semantic Segmentation
Semi-supervised learning for LiDAR semantic segmentation often suffers from error propagation and confirmation bias caused by noisy pseudo-labels. To tackle this chronic issue, we introduce RePL, a novel framework that e…
LIDAR Semantic SegmentationLeveraging Graph Diffusion Models for Network Refinement Tasks
Most real-world networks are noisy and incomplete samples from an unknown target distribution. Refining them by correcting corruptions or inferring unobserved regions typically improves downstream performance. Inspired b…
DenoisingStyle TransferRethinking Reflection in Pre-Training
A language model's ability to reflect on its own reasoning provides a key advantage for solving complex problems. While most recent research has focused on how this ability develops during reinforcement learning, we show…