IBM Research Europe：模拟叠加架构提升 PCM 有效存储密度
Analog Superposition Increases Effective PCM Storage Density (IBM Research Europe)
IBM Research Europe 的研究人员发表了一种模拟相变存储（PCM）架构，通过叠加方式在同一存储介质中存放多个 AI 向量，提升有效存储密度，并可调节精度与密度之间的权衡。在原型 PCM 硬件上的测试显示，推理精度接近全精度软件。该研究以 arXiv 预印本形式发布，采用 CC BY-NC-SA 4.0 许可。
Researchers at IBM Research Europe published “Superposed analog memory for vector storage in neural networks.”
Summary:
Researchers demonstrated an analog phase-change memory architecture that stores multiple AI vectors in superposition, increasing effective storage density with a tunable accuracy tradeoff. Tests on prototype PCM hardware showed inference accuracy close to full-precision software.
Find the technical paper here. October 2026
Vasilopoulos, Athanasios, Patrick Diener, Michael Hersche, Abbas Rahimi, and Abu Sebastian. “Superposed analog memory for vector storage in neural networks.” arXiv preprint arXiv:2610.26118 (2026). CC BY-NC-SA 4.0 License
来源： Semiconductor Engineering 芯片与封装 · semiengineering.com