# Optimizing Few-Step Generation with Adaptive Matching Distillation

arXiv: https://arxiv.org/abs/2602.07345 · HTML: https://mlsanity.com/paper/2602.07345/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2026.

## Built on by (most cited first, 11 of 11)

- [Efficient Video Diffusion Models: Advancements and Challenges](https://mlsanity.com/paper/2604.15911/) (2026)
- [LIVEditor-14B: Lightning Unified Video Editing via In-Context Sparse Attention](https://mlsanity.com/paper/2605.04569/) (2026)
- [Delta Forcing: Trust Region Steering for Interactive Autoregressive Video Generation](https://mlsanity.com/paper/2605.14382/) (2026)
- [FlashMol: High-Quality Molecule Generation in as Few as Four Steps](https://mlsanity.com/paper/2605.07020/) (2026)
- [Qwen-Image-Flash: Rethinking the Training Recipe for Few-Step Distillation](https://mlsanity.com/paper/2606.03746/) (2026)
- [DyMD: Preserving Interaction Dynamics through Distribution Matching Distillation in Few-Step Video World Models](https://mlsanity.com/paper/2609.31349/) (2026)
- [G3-LoRA: Organizing Reward-Weighted Video Data with Gradient-Guided Grouped LoRA](https://mlsanity.com/paper/2609.35189/) (2026)
- [PDMD: Projected Distribution Matching Distillation for Video Diffusion Models](https://mlsanity.com/paper/2609.35768/) (2026)
- [Video Generation Models: A Survey of Post-Training and Alignment](https://mlsanity.com/paper/2610.00812/) (2026)
- [DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation](https://mlsanity.com/paper/2610.02188/) (2026)
- [Golden Anchor for Video Generation](https://mlsanity.com/paper/2610.13217/) (2026)

## Abstract (arXiv)

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.