CASCADE：基于上下文的放松方法用于推测图像解码
CASCADE: Context-Aware Relaxation for Speculative Image Decoding
AI总结：
本文提出CASCADE方法，通过捕捉目标模型隐藏状态中的冗余性，在不额外训练的情况下实现接受放松，提升推测解码效率，实验表明在多模态模型中达到3.6倍加速，保持图像质量和提示一致性。
摘要：
自回归生成是一种强大的高保真图像合成方法，但即使在最先进的加速器上也计算开销大且速度慢。尽管推测解码已被探索以缓解这一瓶颈，现有方法未能实现与文本生成中观察到的效率提升相当的成果。关键限制是图像生成过程中目标模型的高不确定性，导致高草稿令牌拒绝率。本文识别了在基于树的推测解码中自然出现的目标模型行为中被忽视的模式。具体而言，我们正式化了两个属性，语义可交换性和收敛性，这些属性源于目标模型隐藏状态表示中的冗余性。通过在预测令牌树的深度和广度上捕捉这些冗余性，我们的方法在不需额外训练的情况下识别了接受放松的原理性机会。此外，我们通过将目标模型的冗余信号注入drafter训练中，最小化修改来增强独立drafter性能。我们评估了我们的方法在多个文本到图像模型和drafter架构上的表现。结果表明，CASCADE在基于drafter的推测解码中实现了最先进的加速，最高可达3.6倍，同时保持图像质量和文本提示的一致性。
英文摘要：
Autoregressive generation is a powerful approach for high-fidelity image synthesis, but it remains computationally demanding and slow even on the most advanced accelerators. While speculative decoding has been explored to mitigate this bottleneck, existing approaches fail to achieve efficiency gains comparable to those observed in text generation. A key limitation is the target model's high uncertainty during image generation, which leads to high draft token rejection rates. In this work, we identify previously overlooked patterns in the target model's behavior that emerge naturally in tree-based speculative decoding. Specifically, we formalize two properties, semantic interchangeability and convergence, arising from the redundancies in the target model's hidden state representations. By capturing these redundancies across the depth and breadth of the predicted token tree, our method identifies principled opportunities for acceptance relaxation without requiring additional training. Additionally, we enhance standalone drafter performance by injecting the redundancy signals from the target model into drafter training with minimal modification. We evaluate our approach across multiple text-to-image models and drafter architectures. Results show that CASCADE achieves state-of-the-art speedups for drafter-based speculative decoding, with up to 3.6x acceleration, while maintaining image quality and text-prompt fidelity.
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。
