---
id: 20261104-T0-02
title: "神经元就够了？论文质疑AI能力必须存储在突触权重中的假设"
title_en: "Neurons Are All You Need: Paper Challenges Assumption That Learned Capability Resides in Synapses"
url: https://ai.daily.yangsir.net/daily/20261104-T0-02
issue_date: 2026-11-04
publish_date: 2026-11-03T05:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2611.00186
---

# 神经元就够了？论文质疑AI能力必须存储在突触权重中的假设

一篇新论文对人工神经网络沿用数十年的设计假设提出质疑：持久的学习能力是否必须存储在可适应的突触参数中。该假设源于神经生物学对突触连接的认知，并长期决定了网络架构的设计方向。论文探讨了神经元本身作为计算与存储单元的可能性，若成立，将对模型架构和训练方式产生根本性影响。目前该研究为arXiv预印本，尚未经同行评审。

## English Version

**Neurons Are All You Need: Paper Challenges Assumption That Learned Capability Resides in Synapses**

A new paper questions a decades-old design assumption in artificial neural networks: whether durable learned capability must reside exclusively in adaptive synaptic parameters. That assumption stems from neurobiology's understanding of synaptic connectivity and has long shaped network architecture. The paper explores whether neurons themselves can serve as computational and memory units—a finding that, if validated, would fundamentally affect model architecture and training. The work is an arXiv preprint and has not yet been peer-reviewed.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2611.00186)

**详情页**：https://ai.daily.yangsir.net/daily/20261104-T0-02

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*