# SSD：空间推测解码加速自回归图像生成

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-07-14 15:19
- AIHOT 分数：35
- AIHOT 链接：https://aihot.virxact.com/items/cmrkc7blt00g5bizse9grmv7h
- 原文链接：https://x.com/rohanpaul_ai/status/2076929618280661101

## AI 摘要

新方法 Spatially Speculative Decoding (SSD) 通过并行预测图像行，将自回归图像模型加速最高 13.28 倍。SSD 不逐个预测 token，而是利用附近空间信息整行草拟。它添加小型辅助网络，可同时向右和向下查看，预测更易学习的内部特征而非精确视觉 token 标签。主模型并行审查每个草拟块并纠正错误。Janus-Pro、Lumina-mGPT 和 Emu3 的生成速度提升 5.74 至 13.28 倍，基准分数与标准解码相当。Lumina-mGPT 和 Emu3 因更精简的推理循环获得实际加速，Janus-Pro 使用匹配循环。更大 token 网格增益更高，小型训练头可加速过程。

## 正文

Spatially Speculative Decoding （SSD） sped up autoregressive image models up to 13.28X by predicting image rows in parallel.

By drafting whole rows from nearby spatial information， not single tokens.

Most of the time these models take a 2D image and convert it into a long string of tokens. Then they make each token pass through another expensive model.

SSD adds small helper networks that can look right and down . This allows nearby tokens as well as full rows to be drafted at the same time .

Such helpers don't try to predict the exact visual token labels. Instead， they predict more detailed internal features that are easier to learn correctly.

The primary model reviews each drafted block at the same time， correcting any errors so that the entire block is not wasted.

The generation time of Janus-Pro， Lumina-mGPT， and Emu3 was 5.74 to 13.28 times faster， but the benchmark scores were comparable to standard decoding.

Lumina-mGPT and Emu3 saw some wall-clock gains from a leaner inference loop， whereas Janus-Pro used matching loops.

Larger token grids increase the gains but the original model is static and small trained heads speed up the process.

- arxiv. org/abs/2606.20543

Title： "SSD： Spatially Speculative Decoding Accelerates Autoregressive Image Generation"