Recent DiT-based text-to-image models increasingly adopt LLMs as text encoders, yet text conditioning remains largely static and often utilizes only a single LLM layer, despite pronounced semantic hierarchy across LLM layers and non-stationary denoising dynamics over both diffusion time and network depth. To better match the dynamic process of DiT generation and thereby enhance the diffusion model's generative capability, we introduce a unified normalized convex fusion framework equipped with lightweight gates to systematically organize multi-layer LLM hidden states via time-wise, depth-wise, and joint fusion. Experiments establish Depth-wise Semantic Routing as the superior conditioning strategy, consistently improving text-image alignment and compositional generation (e.g., +9.97 on the GenAI-Bench Counting task). Conversely, we find that purely time-wise fusion can paradoxically degrade visual generation fidelity. We attribute this to a train-inference trajectory mismatch: under classifier-free guidance, nominal timesteps fail to track the effective SNR, causing semantically mistimed feature injection during inference. Overall, our results position depth-wise routing as a strong and effective baseline and highlight the critical need for trajectory-aware signals to enable robust time-dependent conditioning.

翻译：近期基于DiT的文本到图像模型越来越多地采用LLM作为文本编码器，然而文本条件处理在很大程度上仍保持静态，且通常仅使用单个LLM层，尽管LLM层间存在显著的语义层次结构，且在扩散时间和网络深度上均呈现非平稳的去噪动态。为更好地匹配DiT生成的动态过程，从而增强扩散模型的生成能力，我们引入了一个配备轻量级门控的统一归一化凸融合框架，通过时间维度、深度维度及联合融合的方式系统化组织多层LLM隐藏状态。实验证明深度语义路由是最优的条件处理策略，能持续提升文本-图像对齐与组合生成能力（例如在GenAI-Bench计数任务中提升+9.97分）。相反，我们发现纯时间维度的融合反而可能降低视觉生成保真度。我们将此归因于训练-推断轨迹失配：在无分类器引导下，名义时间步无法追踪有效信噪比，导致推断过程中出现语义时序错位的特征注入。总体而言，我们的研究将深度路由确立为强大有效的基准方法，并强调轨迹感知信号对于实现鲁棒时间相关条件处理的关键必要性。