SJD++: 改进的推测雅可比解码用于无训练加速离散自回归文本到图像生成
SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image Generation
AI总结：
SJD++通过概率并行解码算法提升自回归文本到图像生成的效率，实现2-3倍的推理延迟降低和2-7倍的步骤压缩，同时保持图像质量。
摘要：
大规模自回归模型可以生成高质量、高分辨率的图像，但生成速度慢，因为这些模型在推理过程中需要数百到数千次顺序前向传递来预测下一个token。为了加速自回归文本到图像生成，我们提出了推测雅可比解码++（SJD++），一种无训练的概率并行解码算法。与传统的下一个token预测不同，SJD++在每次前向传递中进行多token预测，大大减少了生成步骤。具体而言，它整合了来自雅可比解码的迭代多token预测机制，以及来自推测采样中的概率草稿-验证机制。更重要的是，为了进一步加速，SJD++在每次验证阶段后重用高置信度的草稿token，而不是重新采样它们。我们在几个代表性的自回归文本到图像生成模型上进行了广泛的实验，证明SJD++实现了2倍到3倍的推理延迟减少和2倍到7倍的步骤压缩，同时保持视觉质量，没有可观察的降级。
英文摘要：
Large autoregressive models can generate high-quality, high-resolution images but suffer from slow generation speed, because these models require hundreds to thousands of sequential forward passes for next-token prediction during inference. To accelerate autoregressive text-to-image generation, we propose Speculative Jacobi Decoding++ (SJD++), a training-free probabilistic parallel decoding algorithm. Unlike traditional next-token prediction, SJD++ performs multi-token prediction in each forward pass, drastically reducing generation steps. Specifically, it integrates the iterative multi-token prediction mechanism from Jacobi decoding, with the probabilistic drafting-and-verification mechanism from speculative sampling. More importantly, for further acceleration, SJD++ reuses high-confidence draft tokens after each verification phase instead of resampling them all. We conduct extensive experiments on several representative autoregressive text-to-image generation models and demonstrate that SJD++ achieves $2\times$ to $3\times$ inference latency reduction and $2\times$ to $7\times$ step compression, while preserving visual quality with no observable degradation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
