DiFA: Inference-Time Forward-Process Alignment for Diffusion Models
The prevailing inference framework for diffusion models formulates generation fundamentally as a problem of numerical integration. This perspective casts the model as an exact estimator, neglecting the inherent statistical uncertainty of the denoising process. In this work, we propose Forward-Process Aligned Diffusion prediction (DiFA), a training-free framework that reframes inference-time data prediction refinement as a sequential state estimation problem. Rather than reusing past outputs solely for numerical integration, DiFA treats iterative data predictions along the reverse trajectory as correlated observations to build a forward-aligned temporal consensus. Inspired by Kalman filtering, this consensus aggregates historical predictions according to structural consistency and noise-level compatibility. To counteract the over-smoothing tendency of temporal consensus, we introduce a deviation guidance mechanism to adaptively preserve residual details. Empirically, DiFA yields significant improvements on CIFAR-10 and ImageNet across the evaluated metrics, including FID, IS, and FD-DINOv2, demonstrating that aligning inference with the forward statistical structure substantially improves generative fidelity.
Code (2)
Similar Papers 제목 키워드 기반
Look Ma, No GANs! Image Transformation with ModifAE
Existing methods of image to image translation require multiple steps in the training or modification process, and suffer from either an inability to generalize, or long training times. These methods also focus on binary…
Image-to-Image TranslationTranslationLightweight Real-time Makeup Try-on in Mobile Browsers with Tiny CNN Models for Facial Tracking
Recent works on convolutional neural networks (CNNs) for facial alignment have demonstrated unprecedented accuracy on a variety of large, publicly available datasets. However, the developed models are often both cumberso…
Diffusion Models for Generative Outfit Recommendation
Outfit Recommendation (OR) in the fashion domain has evolved through two stages: Pre-defined Outfit Recommendation and Personalized Outfit Composition. However, both stages are constrained by existing fashion products, l…
DifAugGAN: A Practical Diffusion-style Data Augmentation for GAN-based Single Image Super-resolution
It is well known the adversarial optimization of GAN-based image super-resolution (SR) methods makes the preceding SR model generate unpleasant and undesirable artifacts, leading to large distortion. We attribute the cau…
AttributeData AugmentationImage Super-ResolutionSuper-ResolutionDiFaR: Enhancing Multimodal Misinformation Detection with Diverse, Factual, and Relevant Rationales
Generating textual rationales from large vision-language models (LVLMs) to support trainable multimodal misinformation detectors has emerged as a promising paradigm. However, its effectiveness is fundamentally limited by…