Skip to main content We gratefully acknowledge support from the Simons Foundation, member institutions , and all contributors. Donate > cs > arXiv:2407.06023 Help | Advanced Search All fields Title Author Abstract Comments Journal reference ACM classification MSC classification Report number arXiv identifier DOI ORCID arXiv author ID Help pages Full text Search open search GO open navigation menu quick links Login Help Pages About --> Computer Science > Computation and Language arXiv:2407.06023 (cs) [Submitted on 8 Jul 2024 ( v1 ), last revised 24 Jul 2024 (this version, v3)] Title: Distilling System 2 into System 1 Authors: Ping Yu , Jing Xu , Jason Weston , Ilia Kulikov View a PDF of the paper titled Distilling System 2 into System 1, by Ping Yu and 3 other authors View PDF HTML (experimental) Abstract: Large language models (LLMs) can spend extra compute during inference to generate intermediate thoughts, which helps to produce better final responses. Since Chain-of-Thought (Wei et al., 2022), many such System 2 techniques have been proposed such as Rephrase and Respond (Deng et al., 2023a), System 2 Attention (Weston and Sukhbaatar, 2023) and Branch-Solve-Merge (Saha et al., 2023). In this work we investigate self-supervised methods to ``compile'' (distill) higher quality outputs from System 2 techniques back into LLM generations without intermediate reasoning token sequences, as this reasoning has been distilled into System 1. We show that several such techniques can be successfully distilled, resulting in improved results compared to the original System 1 performance, and with less inference cost than System 2. We posit that such System 2 distillation will be an important feature of future continually learning AI systems, enabling them to focus System 2 capabilities on the reasoning tasks that they cannot yet do well. Subjects: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2407.06023 [cs.CL] (or arXiv:2407.06023v3 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2407.06023 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Ping Yu [ view email ] [v1] Mon, 8 Jul 2024 15:17:46 UTC (1,579 KB) [v2] Tue, 9 Jul 2024 16:29:11 UTC (1,579 KB) [v3] Wed, 24 Jul 2024 18:40:36 UTC (595 KB) Full-text links: Access Paper: View a PDF of the paper titled Distilling System 2 into System 1, by Ping Yu and 3 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CL < prev | next > new | recent | 2024-07 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation &times; loading... Data provided by: 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 Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) Links to Code Toggle Papers with Code ( What is Papers with Code? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs 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 . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) About Help contact arXiv Click here to contact arXiv Contact subscribe to arXiv mailings Click here to subscribe Subscribe Copyright Privacy Policy Web Accessibility Assistance arXiv Operational Status