AI Should Sense Better, Not Just Scale Bigger: Adaptive Sensing as a Paradigm Shift

TL;DR

This paper advocates for adaptive sensing in AI, inspired by biological systems, to improve efficiency and robustness, reducing reliance on scaling models and datasets, and addressing environmental and ethical concerns.

Contribution

It introduces adaptive sensing as a paradigm shift in AI, outlining its benefits, challenges, and a research roadmap for integration into various real-world applications.

Findings

Adaptive sensing enables small models to outperform larger ones trained with more data.

Empirical evidence shows improved robustness and efficiency through input-level sensor modulation.

Proposes a comprehensive research agenda including benchmarks and privacy methods.

Abstract

Current AI advances largely rely on scaling neural models and expanding training datasets to achieve generalization and robustness. Despite notable successes, this paradigm incurs significant environmental, economic, and ethical costs, limiting sustainability and equitable access. Inspired by biological sensory systems, where adaptation occurs dynamically at the input (e.g., adjusting pupil size, refocusing vision)--we advocate for adaptive sensing as a necessary and foundational shift. Adaptive sensing proactively modulates sensor parameters (e.g., exposure, sensitivity, multimodal configurations) at the input level, significantly mitigating covariate shifts and improving efficiency. Empirical evidence from recent studies demonstrates that adaptive sensing enables small models (e.g., EfficientNet-B0) to surpass substantially larger models (e.g., OpenCLIP-H) trained with significantly more data and compute. We (i) outline a roadmap for broadly integrating adaptive sensing into real-world applications spanning humanoid, healthcare, autonomous systems, agriculture, and environmental monitoring, (ii) critically assess technical and ethical integration challenges, and (iii) propose targeted research directions, such as standardized benchmarks, real-time adaptive algorithms, multimodal integration, and privacy-preserving methods. Collectively, these efforts aim to transition the AI community toward sustainable, robust, and equitable artificial intelligence systems.