跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 24 天前AI 評分36

IDM(Intrinsic Distribution Matching):內在分佈匹配

Intrinsic Distribution Matching

AI 導讀

IDM(Intrinsic Distribution Matching)在 internal DMD 基礎上加入生成器 EMA 特徵的矩約束,無需外部特徵編碼器。

正文

Hi everyone! We’re excited to share IDM: Intrinsic Distribution Matching.

Can a generator provide both the score supervision and the representation used for distribution matching? Building on internal DMD, IDM retains the full DMD correction and adds an explicit loss on the means and per-coordinate standard deviations of the generator’s own EMA features, without an external feature encoder. A Gaussian-plus-residual decomposition of the reverse-KL field motivates this additional moment constraint.

At 2 NFE, IDM (Z-Image) achieves 94.25% on LongText-CN and 90.20% on LongText-EN, the best scores among the compared methods at this budget. Controlled ablations show consistent gains from the added moment loss, while generator EMA features perform competitively with DINOv2. We evaluate both Z-Image-Turbo 6B and Qwen-Image 20B with full-parameter and LoRA tuning.

Model weights: coming soon. If you find the work interesting, we’d appreciate a GitHub star ⭐ and an HF upvote!

image

image

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