← Back ◬ AI & Machine Learning Mar 30, 2026

MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training

arXiv AI Archived Mar 30, 2026 ✓ Full text saved

arXiv:2603.25813v1 Announce Type: cross Abstract: 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 bi

Full text archived locally

✦ AI Summary · Claude Sonnet

Computer Science > Machine Learning [Submitted on 26 Mar 2026] MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training Yongwan Kim, Sungchul Park 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 this http URL 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). Comments: 20 pages, 4 figures, 8 tables Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2603.25813 [cs.LG] (or arXiv:2603.25813v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2603.25813 Focus to learn more Submission history From: Yongwan Kim [view email] [v1] Thu, 26 Mar 2026 18:21:57 UTC (24 KB) Access Paper: HTML (experimental) view license Current browse context: cs.LG < prev | next > new | recent | 2026-03 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

💬 Team Notes