The Hitchhiker's Guide to System One Models

A practitioner-oriented arXiv guide by Haggai Roitman, submitted 24 October 2026 and revised 29 October 2026, proposes a five-family taxonomy of System 1 decision models — encoder-based classifiers, reward models, non-autoregressive decision engines, contrastive decision models, and extraction models — that map inputs directly to typed, structured decisions in a single forward pass rather than autoregressive token-by-token inference. The guide covers calibration theory including proper scoring rules and Reinforcement Learning for Calibrated Decisions (RLCD), examines IIA violations, ordinal-scale bias, and few-shot recalibration under distribution shift, and describes System 1 + System 2 cascades that route between fast and slow models based on calibrated confidence. It spans 18 application domains, 190 references, over 70 open-source model reproductions, runnable code, a model zoo, and 22 self-assessment questions.

Computer Science Artificial Intelligence Submitted on 24 Oct 2026 v1 https://arxiv.org/abs/2610.33775v1 , last revised 29 Oct 2026 this version, v2 Title:The Hitchhiker's Guide to System 1 Models: Fast Decision-Making Beyond LLMS View PDF https://arxiv.org/pdf/2610.33775 HTML experimental https://arxiv.org/html/2610.33775v2 Abstract:Large Language Models excel at reasoning and generation but their autoregressive, token-by-token inference makes them unnecessarily slow, expensive, and poorly calibrated for the many routine decisions routing, gating, scoring, classification that modern AI systems require at sub-second latency. A growing class of models addresses this gap: System 1 decision models, architectures that map an input directly to a typed, structured decision in a single forward pass. This guide provides a comprehensive, practitioner-oriented treatment of the field. We propose a five-family taxonomy covering encoder-based classifiers, reward models, non-autoregressive decision engines, contrastive decision models, and extraction models, and we provide in-depth coverage of each family's architecture, training methodology, and deployment trade-offs. A central theme is calibration: we develop the theory of proper scoring rules, analyze Reinforcement Learning for Calibrated Decisions RLCD , and examine recently discovered phenomena including IIA violations, ordinal-scale bias, and few-shot recalibration under distribution shift. On the integration side, we describe how System 1 models serve as decision infrastructure within agentic workflows and System 1 + System 2 cascades that route between fast and slow models based on calibrated confidence. The guide covers 18 application domains, maps the rapidly growing open-source ecosystem over 70 model reproductions in weeks , provides runnable code throughout, and includes a practitioner's checklist, an open-source model zoo, and 22 self-assessment questions with detailed answers. Spanning 190 references, it is intended as a unified entry point for researchers and engineers working at the intersection of fast decision-making, calibration, and production AI systems. Submission history From: Haggai Roitman view email https://arxiv.org/show-email/1603e83c/2610.33775 Sat, 24 Oct 2026 16:26:01 UTC 1,262 KB \ v1\ https://arxiv.org/abs/2610.33775v1 v2 Thu, 29 Oct 2026 03:58:10 UTC 1,263 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 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 .