跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-19AI 評分45

SWIFT:透過目標外推實現自我引導的弱至強擴散模型

SWIFT: Self-guided Weak-to-strong dIFfusion models via Target extrapolation

AI 導讀

SWIFT 提出一種自我引導訓練方法,利用凍結的弱去噪器外推回歸目標,改善較強擴散模型的訓練,並降低誤差和訓練成本,無須增加推理開銷。在 ImageNet-256 的一種設定中,SWIFT 以約一半總訓練計算量超越基線最佳 FID;在 RAEv2 上,無推理時引導的 FID 為 1.15,有引導時為 0.99。

正文

Published on Oct 19

Authors:

,

,

,

,

,

Abstract

Swift improves diffusion model training by extrapolating regression targets away from a frozen weak denoiser, reducing error and training cost without inference overhead.

Training high-quality diffusion models is computationally expensive, and inference-time guidance often adds further sampling cost. In this work, we introduce Swift, a simple yet effective self-guided training method that uses a frozen weak denoiser to improve a stronger model without adding inference overhead. For each noisy input, Swift extrapolates the clean regression target away from the weak denoiser's prediction, amplifying components that remain underfit. The weak guide can be an early training checkpoint, requires no backpropagation during main-model training, and is discarded afterward. The resulting model retains the original sampler and supports both conditional and unconditional generation without label dropout or an unconditional branch. We explain this mechanism through a spectral analysis of finite-time population gradient flow in an RKHS. With an early-stopped weak guide, we prove that a nonempty range of extrapolation strengths yields strictly lower posterior-mean error than matched standard training, despite shifting the regression target away from the posterior mean. Experiments on CIFAR-10 and ImageNet-256 demonstrate improvements across model scales and representations. In one ImageNet-256 configuration, Swift surpasses the baseline's best FID using roughly half the total training compute, including guide preparation and training overhead. On RAEv2, it achieves FID 1.15 without inference-time guidance and 0.99 with guidance. These results establish target extrapolation as an effective approach to weak-to-strong diffusion training with efficient single-model inference.

View arXiv page View PDF Project page Add to collection

Get this paper in your agent:

hf papers read 2610.25844

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.25844 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.25844 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.25844 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co