State-Scale MLP Branch Architecture

• State Scale MLP Branch is a neural network pattern that attaches multi-scale output branches at hidden layers to capture both coarse and fine signal components.
• It enables progressive layerwise supervision with a multi-loss framework, facilitating efficient gradient flow and robust feature learning.
• This design improves representation efficiency and performance in tasks like image segmentation and 3D modeling, as demonstrated by T-MLP and PCAS-MLP studies.

A State Scale MLP Branch refers to a neural network architectural pattern that attaches multi-scale output branches—"tails"—at various hidden layers of a Multi-Layer Perceptron (MLP) to enable multi-resolution or level-of-detail (LoD) signal representation and supervision. This approach is prominent in two recent lines of research: (1) Tailed Multi-Layer Perceptron (T-MLP) for LoD signal modeling in implicit neural representations and (2) multi-branch MLP modules, such as PCAS-MLP, for extracting multi-scale features in U-Net-style architectures for image segmentation (Yang et al., 26 Aug 2025, Shi et al., 2023). Such designs facilitate hierarchical supervision and efficient capture of both coarse and fine signal components within a single model.

1. Architectural Principles of State Scale MLP Branches

The central architectural principle is the attachment of dedicated output branches—"tails"—to hidden layers of an MLP backbone. Each tail captures the signal at a distinct effective scale due to the progressively increasing receptive field and abstraction level of deeper layers. In the context of T-MLP, for an MLP backbone of depth $k$ (hidden layers), $k$ tails are appended, with each tail producing an intermediate output $\mathbf{t}_i$ from hidden activation $\mathbf{h}_i$ (Yang et al., 26 Aug 2025). The first tail is linear, while subsequent tails employ a multiplicative elementwise design to facilitate residual learning. Each intermediate output is accumulated to form a sequence of LoD predictions: $$\mathbf{y}_i = \sum_{j=1}^i \mathbf{t}_j,\quad i=1,\dots,k$$ Similarly, in PCAS-MLP within STM-UNet, multiple convolutional branches of varying kernel sizes (1×1, 3×3, 5×5) operate in parallel on intermediate MLP representations, providing a multi-scale response before further processing (Shi et al., 2023).

2. Mathematical Formulations

In T-MLP, the forward computation is defined as follows (Yang et al., 26 Aug 2025):

• Hidden layer activations:

$$\mathbf{h}_0 = \mathbf{x},\quad \mathbf{h}_\ell = \sigma(\mathbf{W}_\ell\,\mathbf{h}_{\ell-1} + \mathbf{b}_\ell),\quad \ell=1\dots k$$

• Output of the first tail:

$$\mathbf{t}_1 = \mathbf{W}^{\text{out}_1}\,\mathbf{h}_1 + \mathbf{b}^{\text{out}_1}$$

• Output of subsequent tails (elementwise multiplication for residuals):

$$\mathbf{t}_\ell = (\mathbf{W}^{\text{out}_{\ell,0}}\,\mathbf{h}_\ell + \mathbf{b}^{\text{out}_{\ell,0}}) \circ (\mathbf{W}^{\text{out}_{\ell,1}}\,\mathbf{h}_\ell + \mathbf{b}^{\text{out}_{\ell,1}}) ,\quad \ell=2\dots k$$

• Progressive output:

$$\mathbf{y}_\ell = \mathbf{y}_{\ell-1} + \mathbf{t}_\ell$$

Where $\circ$ denotes the Hadamard product.

The PCAS-MLP module combines three convolutions in parallel, then processes the result with nonlinearity (GELU), dropout, and serial height/width shifts, before a final channel-wise linear projection and residual connection (Shi et al., 2023).

3. Training and Supervision Strategies

Each tail output $k$0 in T-MLP is trained with explicit supervision against the ground-truth signal or its residual, yielding a layerwise supervised, multi-loss framework: $k$1 The overall loss is

$k$2

where $k$3 are per-tail weights, typically increasing with depth ($k$4, and higher $k$5 for finer predictions).

Progressive training curricula may first freeze all but the shallowest tail, then incrementally activate deeper tails to stabilize coarse-to-fine learning. This approach alleviates vanishing gradients and forces each layer to encode semantically meaningful multi-scale features (Yang et al., 26 Aug 2025).

4. Practical Implementations and Hyperparameters

Practical instantiations of State Scale MLP Branch architectures differ by application domain:

• T-MLP for LoD Signal Representation (Yang et al., 26 Aug 2025):
  • MLP backbone width (per layer): $k$6 (e.g., 256).
  • Nonlinearities: ReLU or sine (as used in SIREN).
  • Number of tails: $k$7 (number of hidden layers).
  • Output heads: linear for first tail; elementwise-multiplicative for others.
  • Loss weights: example $k$8 for $k$9.
  • Optimizer: Adam, initial LR $\mathbf{t}_i$0–$\mathbf{t}_i$1, decayed at 0.25× schedule.
  • Data: Random pixel sampling for images, surface-oriented sampling for SDFs.
• PCAS-MLP for STM-UNet (Shi et al., 2023):
  • Operates at C=512 channels in the bottleneck.
  • Parallel convolutions: 1×1, 3×3 (pad=1), 5×5 (pad=2).
  • Axial shifting blocks $\mathbf{t}_i$2 (e.g., 16), with uniform shift offsets.
  • GELU activation, Dropout $\mathbf{t}_i$3 at two points.
  • Optimizer: Adam, LR $\mathbf{t}_i$4.
  • U-Net stages: channels [32, 64, 128, 256, 512].
  • Training: 300 epochs, batch size 8, 512×512 image size.

5. Comparative Performance and Applications

State Scale MLP Branch designs deliver substantial empirical gains in representation efficiency, quality, and speed, especially for LoD tasks.

Model/Task	Params	3D Shape CD$\mathbf{t}_i$5	3D Normal Cons$\mathbf{t}_i$6	Image PSNR$\mathbf{t}_i$7	ISIC-2018 mIoU$\mathbf{t}_i$8
T-MLP (Yang et al., 26 Aug 2025)	∼266k	1.513	98.03%	30.63	–
SIREN	∼265k	1.769	–	28.02	–
NGLOD	1.35M	1.975	–	–	–
BACON	264k	1.787	–	–	–
BANF	2.08M	4.683	–	–	–
Baseline U-NeXt-L	3.99M	–	–	–	0.7720
STM-UNet	6.12M	–	–	–	0.7984

T-MLP demonstrates lower Chamfer distance and higher normal consistency on 3D modeling benchmarks, outperforming SIREN, NGLOD, BACON, and BANF at comparable or lower parameter counts. For 2D image regression (DIV2K), T-MLP achieves a PSNR gain of over 2 dB versus SIREN. Training times are an order of magnitude faster than comparable LoD approaches.

PCAS-MLP, as integrated into STM-UNet, improves mIoU and mDice by approximately +2.6% and +1.1% over AS-MLP baselines, with smaller model size increases compared to pure CNN or hybrid Transformer models (Shi et al., 2023).

6. Significance and Broader Implications

State Scale MLP Branch architectures enable a single network to efficiently encode a continuous, explicit level-of-detail hierarchy. Shallow tails capture low-frequency coarse structure, while deeper branches add high-frequency detail. Layerwise supervision provides effective gradient propagation, mitigating vanishing gradient issues typical of deep MLPs. For practical systems, progressive output can support real-time coarse-to-fine rendering or streaming.

A plausible implication is that these multi-scale branching MLP techniques may inform the design of future signal representation models and efficient vision backbones by providing robust layerwise decodability and multi-scale feature extraction in compact, end-to-end trainable frameworks.

7. Related Approaches and Future Directions

While State Scale MLP Branches as described in T-MLP and PCAS-MLP are specific to LoD and segmentation domains, the principle of layerwise multi-scale branching is extensible to other tasks requiring explicit scale hierarchies, including neural rendering, adaptive transmission, and multi-resolution generative modeling. The use of elementwise-multiplicative branches (T-MLP) and parallel axial-shifted filtering (PCAS-MLP) suggests a convergence between efficient MLP-based models and convolutional/attention counterparts. Future directions include scaling to larger models, integration with attention, and exploration of dynamic per-branch supervision schedules to further enhance adaptation to varying task requirements (Yang et al., 26 Aug 2025, Shi et al., 2023).