AsyncDiff: 异步时间步条件化以提升文本到图像扩散推理
AsyncDiff: Asynchronous Timestep Conditioning for Enhanced Text-to-Image Diffusion Inference
AI总结：
AsyncDiff通过异步时间步条件化提升文本到图像扩散推理性能，采用轻量时间步预测模块和动态缩放参数实现更灵活的噪声控制。
摘要：
文本到图像扩散推理通常遵循同步计划，其中数值积分器将潜在状态推进到与去噪器条件化相同的时间步。我们提出了一种异步推理机制，将这两个过程解耦，允许去噪器在不同的、学习的时间步上进行条件化，同时保持图像更新计划不变。一个轻量的时间步预测模块（TPM），通过组相对策略优化（GRPO）进行训练，根据当前状态选择更可行的条件化时间步，有效地选择一个期望的噪声水平以控制图像细节和纹理丰富性。在部署时，可以使用一个缩放超参数来插值原始和去同步时间步之间，从而实现保守或激进的调整。为了保持研究的计算成本低廉，我们限制推理在SD3.5上为15步，在Flux上为10步。在Stable Diffusion 3.5 Medium和Flux.1-dev上评估MS-COCO 2014和T2I-CompBench数据集，我们的方法优化了一个综合奖励，平均图像奖励、HPSv2、CLIP分数和Pick分数，并显示出一致的改进。
英文摘要：
Text-to-image diffusion inference typically follows synchronized schedules, where the numerical integrator advances the latent state to the same timestep at which the denoiser is conditioned. We propose an asynchronous inference mechanism that decouples these two, allowing the denoiser to be conditioned at a different, learned timestep while keeping image update schedule unchanged. A lightweight timestep prediction module (TPM), trained with Group Relative Policy Optimization (GRPO), selects a more feasible conditioning timestep based on the current state, effectively choosing a desired noise level to control image detail and textural richness. At deployment, a scaling hyper-parameter can be used to interpolate between the original and de-synchronized timesteps, enabling conservative or aggressive adjustments. To keep the study computationally affordable, we cap the inference at 15 steps for SD3.5 and 10 steps for Flux. Evaluated on Stable Diffusion 3.5 Medium and Flux.1-dev across MS-COCO 2014 and T2I-CompBench datasets, our method optimizes a composite reward that averages Image Reward, HPSv2, CLIP Score and Pick Score, and shows consistent improvement.
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。
