IBM Quantum has led significant advancements in quantum hardware, scaling from 5-qubit Canary processors to the record-breaking 1,121-qubit Condor, pushing the boundaries of practical quantum computing.
Qibocal is an open-source software library that provides tools and protocols for calibrating and characterizing superconducting quantum devices within the Qibo framework.
A novel Binary Search ADC design that significantly reduces area and power overhead compared to existing solutions, combined with an in-training optimization approach to further minimize ADC complexity with negligible impact on accuracy, enabling efficient and compact flexible electronics devices.
A novel analog in-memory computing architecture based on gain cell memories can perform attention computations for large language models with significantly lower latency and energy consumption compared to GPUs.
유연한 폴리이미드 기판에 제작된 인듐 갈륨 아연 산화물 박막 트랜지스터 기반의 32비트 RISC-V 마이크로프로세서 Flex-RV가 소개되었다. Flex-RV는 내장형 기계 학습 가속기를 포함하며, 기존 RISC-V 명령어 세트를 확장하여 기계 학습 워크로드를 실행할 수 있다.
Flex-RV, a 32-bit RISC-V microprocessor fabricated with indium gallium zinc oxide thin-film transistors on a flexible polyimide substrate, enables an ultralow-cost bendable microprocessor with integrated machine learning hardware accelerator.
The Huawei Mate XT is a tri-fold smartphone that offers a large 10.2-inch tablet-sized display when unfolded, while maintaining a relatively compact form factor when folded.
This paper introduces a CMOS-based analog spiking neuron circuit that utilizes time-domain information, such as time interval and pulse width, to construct a hardware-friendly physical reservoir computing system.
실험실 기반 기술과 동등한 성능과 수율을 가진 300mm CMOS 파일럿 라인에서 제조된 초전도 트랜지스터 큐비트
Demonstration of manufacturing high-coherence superconducting transmon qubits using advanced industrial 300 mm CMOS fabrication processes, enabling scalable quantum computing.