Optimizing Few-Step Generation with Adaptive Matching Distillation

TL;DR

This paper introduces Adaptive Matching Distillation (AMD), a novel framework that improves the stability and performance of few-step generative models by explicitly avoiding unreliable regions during training.

Contribution

The paper proposes AMD, a self-correcting distillation method that detects and escapes Forbidden Zones, enhancing sample quality and training robustness in generative tasks.

Findings

AMD improves HPSv2 score on SDXL from 30.64 to 31.25

Enhances sample fidelity and robustness across image and video tasks

Outperforms state-of-the-art baselines in benchmarks

Abstract

Distribution Matching Distillation (DMD) is a powerful acceleration paradigm, yet its stability is often compromised in Forbidden Zone, regions where the real teacher provides unreliable guidance while the fake teacher exerts insufficient repulsive force. In this work, we propose a unified optimization framework that reinterprets prior art as implicit strategies to avoid these corrupted regions. Based on this insight, we introduce Adaptive Matching Distillation (AMD), a self-correcting mechanism that utilizes reward proxies to explicitly detect and escape Forbidden Zones. AMD dynamically prioritizes corrective gradients via structural signal decomposition and introduces Repulsive Landscape Sharpening to enforce steep energy barriers against failure mode collapse. Extensive experiments across image and video generation tasks (e.g., SDXL, Wan2.1) and rigorous benchmarks (e.g., VBench, GenEval) demonstrate that AMD significantly enhances sample fidelity and training robustness. For instance, AMD improves the HPSv2 score on SDXL from 30.64 to 31.25, outperforming state-of-the-art baselines. These findings validate that explicitly rectifying optimization trajectories within Forbidden Zones is essential for pushing the performance ceiling of few-step generative models.