CVOct 24, 2025

Improved Training Technique for Shortcut Models

arXiv:2510.21250v13 citationsh-index: 4
Originality Highly original
AI Analysis

It addresses critical limitations for researchers and practitioners in generative AI, making shortcut models more viable and competitive, though it is incremental as it builds on an existing paradigm.

This paper tackled performance bottlenecks in shortcut models for generative modeling, introducing a unified training framework (iSM) that resolved five core issues, leading to substantial FID improvements on ImageNet 256x256 across one-step, few-step, and multi-step generation.

Shortcut models represent a promising, non-adversarial paradigm for generative modeling, uniquely supporting one-step, few-step, and multi-step sampling from a single trained network. However, their widespread adoption has been stymied by critical performance bottlenecks. This paper tackles the five core issues that held shortcut models back: (1) the hidden flaw of compounding guidance, which we are the first to formalize, causing severe image artifacts; (2) inflexible fixed guidance that restricts inference-time control; (3) a pervasive frequency bias driven by a reliance on low-level distances in the direct domain, which biases reconstructions toward low frequencies; (4) divergent self-consistency arising from a conflict with EMA training; and (5) curvy flow trajectories that impede convergence. To address these challenges, we introduce iSM, a unified training framework that systematically resolves each limitation. Our framework is built on four key improvements: Intrinsic Guidance provides explicit, dynamic control over guidance strength, resolving both compounding guidance and inflexibility. A Multi-Level Wavelet Loss mitigates frequency bias to restore high-frequency details. Scaling Optimal Transport (sOT) reduces training variance and learns straighter, more stable generative paths. Finally, a Twin EMA strategy reconciles training stability with self-consistency. Extensive experiments on ImageNet 256 x 256 demonstrate that our approach yields substantial FID improvements over baseline shortcut models across one-step, few-step, and multi-step generation, making shortcut models a viable and competitive class of generative models.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes