Figure 1. Motivation. (a) Dense all-in-one restoration meth-
ods [24, 57] often inefficiently allocate parameters when handling
multiple degradation types. (b) While recent Mixture-of-Experts
(MoE) approaches [61, 62] address this through sparse computa-
tion, their rigid routing mechanisms uniformly distribute inputs
across experts without considering the natural relationships be-
tween degradations. (c) To overcome these limitations, we intro-
duce Complexity Experts - adaptive processing blocks with size-
varying computational units. Our framework dynamically allocates
model capacity using a spring-inspired force mechanism that con-
tinuously guides routing decisions toward simpler experts when
possible, with the force proportional to the complexity of the input
degradation. While initially designed for computational efficiency,
this approach naturally emerges as a task-discriminative learning
framework, assigning degradations to the most suitable experts.
This makes it particularly effective for all-in-one restoration meth-
ods, where both task-specific processing and cross-degradation
knowledge sharing are crucial.