CAFE: Channel-Autoregressive Factorized Encoding for Robust Biosignal Spatial Super-Resolution

TL;DR

CAFE, a plug-and-play rollout generation scheme that reconstructs the full montage in geometry-aligned stages, demonstrates plug-and-play generality across $3$ backbones and achieves consistently better reconstruction than $5$ representative baselines.

Abstract

High-density biosignal recordings are critical for neural decoding and clinical monitoring, yet real-world deployments often rely on low-density (LD) montages due to hardware and operational constraints. This motivates spatial super-resolution from LD observations, but heterogeneous dependencies under sparse and noisy measurements often lead to artifact propagation and false non-local correlations. To address this, we propose CAFE, a plug-and-play rollout generation scheme that reconstructs the full montage in geometry-aligned stages. Starting from the LD channels, CAFE first recovers nearby channels and then progressively expands to more distal regions, exploiting reliable local structure before introducing non-local interactions. During training, step-wise supervision is applied over channel groups and teacher forcing with epoch-level scheduled sampling along the group dimension is utilized to reduce exposure bias, enabling parallel computation across steps. At test time, CAFE performs an autoregressive rollout across groups, while remaining plug-and-play by reusing any temporal backbone as the shared predictor. Evaluated on $4$ modalities and $6$ datasets, CAFE demonstrates plug-and-play generality across $3$ backbones (MLP, Conv, Transformer) and achieves consistently better reconstruction than $5$ representative baselines.

Figures (8)

• Figure 1: Motivation: heterogeneous channel topology. Top: chord diagrams illustrating that cross-channel dependencies (shown on EEG as an example) are heterogeneous across sensors, with stronger interactions concentrated within specific functional regions. Bottom: schematic comparison of channel-handling strategies. Colored blocks denote channels and their features; square blocks are features before applying a strategy, and rounded blocks are the resulting representations.
• Figure 2: Performance of CAFE. Results (SNR) are averaged from all SR scale factors. CAFE outperforms strong baselines in 6 commonly used datasets.
• Figure 3: CAFE model overview.
• Figure 4: Effect of decoding order in group-wise autoregressive reconstruction.
• Figure 5: Effect of rollout granularity (channels per step) on SEED.