用自研系统自动训练专家级语言模型，还能在普通电脑上跑。
MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training
- 自研研究流水线自动完成数据生成与调参迭代
- 模型在普通电脑上实现推理，准确率最高达98.51%
- 适合想低成本部署专业模型的开发者
我们提出MAGNET（Model Autonomously Growing Network），一个去中心化系统，可在通用硬件上自主生成、训练和部署领域专家语言模型。系统包含四个组件：(1) 自动化机器学习研究流程，实现数据生成、超参数探索、评估与错误驱动迭代；(2) BitNet b1.58 三值训练，通过bitnet.cpp实现纯CPU推理，无需GPU；(3) 基于DiLoCo的分布式合并，高效聚合领域专家模型；(4) 在HOOTi EVM链上追踪贡献。通过三个案例验证：视频安全分类（平衡准确率0.9287至0.9851）、加密货币方向预测（命中率41%至54.9%）、BitNet超参数优化（10阶段扫描，验证损失降低16.7%）。
原文摘要 · Abstract (English)
We present MAGNET (Model Autonomously Growing Network), a decentralized system for autonomous generation, training, and serving of domain-expert language models across commodity hardware. MAGNET integrates four components: (1) autoresearch, an autonomous ML research pipeline that automates dataset generation, hyperparameter exploration, evaluation, and error-driven iteration; (2) BitNet b1.58 ternary training, enabling CPU-native inference via bitnet.cpp without GPU hardware; (3) DiLoCo-based distributed merging for communication-efficient aggregation of domain specialists; and (4) on-chain contribution tracking on the HOOTi EVM chain. We validate autoresearch through three case studies: video safety classification (balanced accuracy 0.9287 to 0.9851), cryptocurrency directional prediction (41% to 54.9% hit rate), and BitNet hyperparameter optimization (10-phase sweep, -16.7% validation loss).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。