Answers are grounded in this paper and cite their sources.
Demonstrates a reconfigurable device's versatility in sensing and processing for optical applications, suggesting a path for advanced vision systems.
Integrating sensing and processing capabilities into a single device offers a promising route to energy‐efficient artificial vision. While multi‐mode devices have shown promise in the visible range, extending these functionalities to the short‐wave infrared region remains elusive. Moreover, achieving dynamically tunable spectral responsivity to adapt to complex environments remains challenging. Here, we report a reconfigurable heterostructure device based on carbon nanotubes and ferroelectric semiconductor α ‐In 2 Se 3 . The device integrates three switchable functionalities: broadband sensing, spectral selective photodetection, and optoelectronic synapse. By leveraging the ferroelectric polarization and bias‐modulated band alignment, the device can be seamlessly switched among these modes. In self‐powered broadband photodetection mode, it exhibits an ultra‐broadband response (325–2100 nm) at room temperature with a specific detectivity of 5.16 × 10 9 Jones and responsivity of 205.3 mA/W at 1800 nm. In tunable spectral selectivity mode, the device can be switched between multi‐band collaborative detection and strong‐light interference filtering detection by applying different bias voltages. In the optoelectronic synaptic plasticity mode, comprehensive synaptic plasticity is demonstrated. To demonstrate this trimodal versatility, we implemented a logic gate, Morse code encoding, and a reservoir computing system for handwritten digit recognition. This work provides a robust “sensing‐preprocessing‐computing” platform for future intelligent vision systems.
No takes yet. Share an insight, caveat, or question.
Zhou et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: