跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 24 天前AI 評分47

解鎖預訓練自回歸圖像模型的任意順序生成能力

Unlocking Any-Order Generation in Pretrained Autoregressive Image Models

AI 導讀

研究提出以未來感知位置嵌入和排序課程,將預訓練的光柵掃描自回歸圖像模型改造成任意順序生成器,支援修補、外延和編輯。方法在 ImageNet 修補上的 FID 為 6.21,優於 MaskGIT 的 10.44;額外計算量為預訓練以外的 13-33%,並保留標準自回歸推理架構。

正文

Published on Oct 29

Authors:

,

Abstract

Pretrained raster-scan autoregressive image models can be adapted into flexible any-order generators via future-aware positional embeddings and an ordering curriculum, enabling inpainting and outpainting with minimal extra compute.

Autoregressive (AR) image generators are powerful in principle but inflexible in practice: training with a fixed raster-scan order hardcodes a single factorization of the image distribution into the model, leaving it unable to inpaint, outpaint, or edit at inference time. Existing solutions, including the modern discrete-diffusion family, train from scratch on a mixture of token orderings, paying roughly 3x the compute of standard raster-scan training; discrete diffusion further sacrifices compatibility with the AR inference ecosystem (KV caching, vLLM) by construction. We argue that this cost is unnecessary. A pretrained raster-scan AR model has already learned the joint distribution over image tokens; what it lacks is not knowledge but the ability to express that knowledge through non-sequential conditional pathways. We propose a recipe for adapting pretrained raster-scan models into any-order generators using two components: future-aware positional embeddings that resolve a target-position identifiability problem at negligible overhead, and an ordering curriculum that traces a stable optimization path from the pretrained solution to a good any-order basin. Applied to four base models across two architectures and two scales (LlamaGen-L/XL, RAR-L/XL), the recipe substantially outperforms MaskGIT on ImageNet inpainting (FID 6.21 vs. 10.44) and outpainting, is competitive with MAR-L, trades a 1.2-1.9 increase in class-conditional FID for this flexibility, and costs 13-33% additional compute beyond pretraining, while leaving the standard AR inference stack untouched. Inpainting, outpainting, and editing, our results suggest, need not be added to autoregressive image models. They can be unlocked from the ones that exist. Code is available at https://github.com/Ordered-AR/unlockedadaptation.

View arXiv page View PDF GitHub 0 Add to collection

Get this paper in your agent:

hf papers read 2610.39293

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/2610.39293 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/2610.39293 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/2610.39293 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