SOURCE-LINKED INTELLIGENCE

FORGE: Forward-Only Test-Time Adaptation for Integer-Only Vision Models on Microcontrollers

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Vision models deployed on microcontrollers (MCUs) are quantized to integer-only arithmetic and run in inference-only runtimes that do not carry the machinery backpropagation needs: the standard tool for adapting a model to the distribution shift (sensor noise, blur, lighting) it meets in the field. Existing forward-only test-time adaptation (TTA) methods either run only on server- or edge-GPU-class models (not true microcontroller integer execution), or require the batch-normalization (BN) layers that integer deployment fuses away. We present a forward-only TTA method that operates on deployed

Read original source ↗ Open in workspace

recordType

paper

region

Global