Skip to main content We gratefully acknowledge support from the Simons Foundation, member institutions , and all contributors. Donate > cs > arXiv:2411.07279 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 > Artificial Intelligence arXiv:2411.07279 (cs) [Submitted on 11 Nov 2024 ( v1 ), last revised 25 Mar 2025 (this version, v2)] Title: The Surprising Effectiveness of Test-Time Training for Few-Shot Learning Authors: Ekin Akyürek , Mehul Damani , Adam Zweiger , Linlu Qiu , Han Guo , Jyothish Pari , Yoon Kim , Jacob Andreas View a PDF of the paper titled The Surprising Effectiveness of Test-Time Training for Few-Shot Learning, by Ekin Aky\"urek and 7 other authors View PDF HTML (experimental) Abstract: Language models (LMs) have shown impressive performance on tasks within their training distribution, but often struggle with structurally novel tasks even when given a small number of in-context task examples. We investigate the effectiveness of test-time training (TTT) -- temporarily updating model parameters during inference using a loss derived from input data -- as a mechanism for improving LMs' reasoning and few-shot learning capabilities. On the Abstraction and Reasoning Corpus (ARC), performing TTT with in-context examples yields up to $6\times$ higher accuracy compared to fine-tuned baselines -- reaching $53.0\%$ on the public validation set with an 8B-parameter LM and $61.9\%$ when ensembled with program-synthesis methods, matching average human performance. On BIG-Bench Hard (BBH), TTT on in-context examples surpasses standard few-shot prompting in the $10$-shot setting by $7.3$ percentage points ($50.5\%$ to $57.8\%$). Our findings highlight the limitations of in-context learning for novel tasks and demonstrate the potential of test-time training to enhance language model adaptability. Comments: Preprint Subjects: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG) Cite as: arXiv:2411.07279 [cs.AI] (or arXiv:2411.07279v2 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2411.07279 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Ekin Akyürek [ view email ] [v1] Mon, 11 Nov 2024 18:59:45 UTC (2,622 KB) [v2] Tue, 25 Mar 2025 03:36:21 UTC (3,340 KB) Full-text links: Access Paper: View a PDF of the paper titled The Surprising Effectiveness of Test-Time Training for Few-Shot Learning, by Ekin Aky\"urek and 7 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.AI < prev | next > new | recent | 2024-11 Change to browse by: cs cs.CL cs.LG 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