Neural Computers

A paper posted to arXiv on 7 April 2026 (revised 16 April 2026, v2) by Mingchen Zhuge and co-authors proposes "Neural Computers" (NCs), a machine form that unifies computation, memory, and I/O in a learned runtime state, with the long-term goal of a general-purpose "Completely Neural Computer" (CNC). The authors instantiate NCs as video models that roll out screen frames from instructions, pixels, and user actions in CLI and GUI settings, and report that NCs acquire elementary interface primitives, especially I/O alignment and short-horizon control, while routine reuse, controlled updates, and symbolic stability remain challenging. The paper outlines a roadmap toward CNCs, positioning the work as a computing paradigm beyond today's agents and conventional computers.

Computer Science Machine Learning Submitted on 7 Apr 2026 v1 https://arxiv.org/abs/2604.06425v1 , last revised 16 Apr 2026 this version, v2 Title:Neural Computers View PDF https://arxiv.org/pdf/2604.06425 HTML experimental https://arxiv.org/html/2604.06425v2 Abstract:We propose a new frontier: Neural Computers NCs that unify computation, memory, and I/O of traditional computers in a learned runtime state. Our long-term goal is the Completely Neural Computer CNC : the mature, general-purpose realization of this emerging machine form, with stable execution, explicit reprogramming, and durable capability reuse. As an initial step, we study whether elementary NC primitives can be learned solely from collected I/O traces, without instrumented program state. Concretely, we instantiate NCs as video models that roll out screen frames from instructions, pixels, and user actions when available in CLI and GUI settings. We show that NCs can acquire elementary interface primitives, especially I/O alignment and short-horizon control, while routine reuse, controlled updates, and symbolic stability remain challenging. We outline a roadmap toward CNCs, to establish a new computing paradigm beyond today's agents and conventional computers. Submission history From: Mingchen Zhuge view email https://arxiv.org/show-email/e741ae62/2604.06425 Tue, 7 Apr 2026 20:01:05 UTC 27,484 KB \ v1\ https://arxiv.org/abs/2604.06425v1 v2 Thu, 16 Apr 2026 21:05:33 UTC 27,474 KB Additional Features References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender IArxiv Recommender What is IArxiv? https://iarxiv.org/about arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .