Real-world images often suffer from spatially diverse degradations such as haze, rain, snow, and low-light, significantly impacting visual quality and downstream vision tasks. Existing all-in-one restoration (AIR) approaches either depend on external text prompts or embed hand-crafted architectural priors (e.g., frequency heuristics); both impose discrete, brittle assumptions that weaken generalization to unseen or mixed degradations. To address this limitation, we propose to reframe AIR as learned latent prior inference, where degradation-aware representations are automatically inferred from the input without explicit task cues. Based on latent priors, we formulate AIR as a structured reasoning paradigm: (1) which features to route (adaptive feature selection), (2) where to restore (spatial localization), and (3) what to restore (degradation semantics). We design a lightweight decoding module that efficiently leverages these latent encoded cues for spatially-adaptive restoration. Extensive experiments across six common degradation tasks, five compound settings, and previously unseen degradations demonstrate that our method outperforms state-of-the-art (SOTA) approaches, achieving an average PSNR improvement of 1.68 dB while being three times more efficient.

翻译：现实世界图像常遭受空间多样的退化影响，如雾霾、雨雪和低光照，这些退化显著降低了视觉质量并影响下游视觉任务。现有的一体化复原方法要么依赖外部文本提示，要么嵌入手工设计的结构先验（如频率启发式）；这两种方式均引入了离散且脆弱的假设，削弱了对未见或混合退化的泛化能力。为克服这一局限，我们提出将一体化复原重构为学习型潜在先验推断框架，其中退化感知表征可直接从输入中自动推断，无需显式任务提示。基于潜在先验，我们将一体化复原构建为结构化推理范式：（1）路由哪些特征（自适应特征选择），（2）在何处复原（空间定位），以及（3）复原何种内容（退化语义）。我们设计了一个轻量级解码模块，能高效利用这些潜在编码线索实现空间自适应复原。在六种常见退化任务、五种复合场景及未见退化类型上的大量实验表明，本方法优于当前最先进技术，平均PSNR提升达1.68 dB，同时计算效率提升三倍。