.png){kind=icon}
.png){kind=icon}
Return
Cover Image
.png){kind=icon}
OA
AI
DOI：10.1002/inf2.12588.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In2Se3-based ferroelectric artificial neurons with a multi-terminal electrode architecture have been developed to directly encode and process spatiotemporal signals at the device level. The resistive response of the In2Se3 ferroelectric memristor is harnessed for spatiotemporal-specific neuronal behavior by engineering inherent in-plane ferroelectricity and exploiting the broad regulation capability of a coplanar multi-electrode configuration, enabling adaptable on-device information processing and culminating in a speech recognition proof-of-concept with accuracy approaching 100%. These results demonstrate the device's promise as a compact, energy-efficient building block for future on-chip neuromorphic studies.
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.