{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/tttflow-unsupervised-test-time-training-with","title":"TTTFlow: Unsupervised Test-Time Training with Normalizing Flow","arxiv_id":"2210.11389","date":"2022-10-20","proceeding":null,"authors":["David Osowiechi","Gustavo A. Vargas Hakim","Mehrdad Noori","Milad Cheraghalikhani","Ismail Ben Ayed","Christian Desrosiers"],"abstract":"A major problem of deep neural networks for image classification is their vulnerability to domain changes at test-time. Recent methods have proposed to address this problem with test-time training (TTT), where a two-branch model is trained to learn a main classification task and also a self-supervised task used to perform test-time adaptation. However, these techniques require defining a proxy task specific to the target application. To tackle this limitation, we propose TTTFlow: a Y-shaped architecture using an unsupervised head based on Normalizing Flows to learn the normal distribution of latent features and detect domain shifts in test examples. At inference, keeping the unsupervised head fixed, we adapt the model to domain-shifted examples by maximizing the log likelihood of the Normalizing Flow. Our results show that our method can significantly improve the accuracy with respect to previous works.","url_abs":"https://arxiv.org/abs/2210.11389v1","url_pdf":"https://arxiv.org/pdf/2210.11389v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"tttflow-unsupervised-test-time-training-with","repo_url":"https://github.com/gustavovargashakim/tttflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"source-free-domain-adaptation","task_name":"Source-Free Domain Adaptation"},{"task_slug":"test-time-adaptation","task_name":"Test-time Adaptation"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"normalizing-flows","method_name":"Normalizing Flows"},{"method_slug":"test","method_name":"Test"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2210.11389","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}