{"url":"/task/source-free-domain-adaptation","name":"Source-Free Domain Adaptation","slug":"source-free-domain-adaptation","description_markdown":"Source-Free Domain Adaptation (SFDA) is a domain adaptation method in machine learning and computer vision where the goal is to adapt a pre-trained model to a new, target domain without access to the source domain data. This approach is advantageous in scenarios where sharing the source data is impractical due to privacy concerns, data size, or proprietary restrictions","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":188,"papers_with_code":103,"benchmarks":7,"benchmark_tables_in_archive":7,"benchmark_tables_shown":7,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":6,"subtasks":2,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/source-free-domain-adaptation-on-visda-2017","slug":"source-free-domain-adaptation-on-visda-2017","dataset":"VisDA-2017","dataset_url":"/dataset/visda-2017","rows_in_archive":10,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"RCL","paper_title":"Empowering Source-Free Domain Adaptation with MLLM-driven Curriculum Learning","paper_url":"/paper/empowering-source-free-domain-adaptation-with","paper_date":"2024-05-28","arxiv_id":"2405.18376","code_links":[{"title":"Dong-Jie-Chen/RCL","url":"https://github.com/Dong-Jie-Chen/RCL"}],"syntology":null}},{"leaderboard":"/sota/source-free-domain-adaptation-on-pacs","slug":"source-free-domain-adaptation-on-pacs","dataset":"PACS","dataset_url":"/dataset/pacs","rows_in_archive":3,"metrics":["Average Accuracy"],"first_row_in_archive_order":{"model":"SPM","paper_title":"Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation","paper_url":"/paper/shuffle-patchmix-augmentation-with-confidence","paper_date":"2025-05-30","arxiv_id":"2505.24216","code_links":[{"title":"PrasannaPulakurthi/SPM","url":"https://github.com/PrasannaPulakurthi/SPM"}],"syntology":null}},{"leaderboard":"/sota/source-free-domain-adaptation-on-cityscapes","slug":"source-free-domain-adaptation-on-cityscapes","dataset":"Cityscapes to ACDC","dataset_url":"/dataset/acdc-adverse-conditions-dataset-with","rows_in_archive":2,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"CMA","paper_title":"Contrastive Model Adaptation for Cross-Condition Robustness in Semantic Segmentation","paper_url":"/paper/contrastive-model-adaptation-for-cross","paper_date":"2023-03-09","arxiv_id":"2303.05194","code_links":[{"title":"brdav/cma","url":"https://github.com/brdav/cma"}],"syntology":{"n":6,"n_ran":5,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/source-free-domain-adaptation-on-cityscapes-1","slug":"source-free-domain-adaptation-on-cityscapes-1","dataset":"Cityscapes to Dark Zurich","dataset_url":"/dataset/dark-zurich","rows_in_archive":1,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"CMA","paper_title":"Contrastive Model Adaptation for Cross-Condition Robustness in Semantic Segmentation","paper_url":"/paper/contrastive-model-adaptation-for-cross","paper_date":"2023-03-09","arxiv_id":"2303.05194","code_links":[{"title":"brdav/cma","url":"https://github.com/brdav/cma"}],"syntology":{"n":6,"n_ran":5,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/source-free-domain-adaptation-on-gta5-to","slug":"source-free-domain-adaptation-on-gta5-to","dataset":"GTA5 to Cityscapes","dataset_url":"/dataset/gta5","rows_in_archive":1,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"HALO","paper_title":"Hyperbolic Active Learning for Semantic Segmentation under Domain Shift","paper_url":"/paper/hyperbolic-active-learning-for-semantic","paper_date":"2023-06-19","arxiv_id":"2306.11180","code_links":[{"title":"paolomandica/HALO","url":"https://github.com/paolomandica/HALO"}],"syntology":{"n":8,"n_ran":3,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/source-free-domain-adaptation-on-synthia-to","slug":"source-free-domain-adaptation-on-synthia-to","dataset":"SYNTHIA-to-Cityscapes","dataset_url":"/dataset/synthia","rows_in_archive":1,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"STPL","paper_title":"Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free Domain Adaptation for Video Semantic Segmentation","paper_url":"/paper/spatio-temporal-pixel-level-contrastive","paper_date":"2023-03-25","arxiv_id":"2303.14361","code_links":[{"title":"shaoyuanlo/stpl","url":"https://github.com/shaoyuanlo/stpl"}],"syntology":null}},{"leaderboard":"/sota/source-free-domain-adaptation-on-viper-to","slug":"source-free-domain-adaptation-on-viper-to","dataset":"VIPER-to-Cityscapes","dataset_url":null,"rows_in_archive":1,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"STPL","paper_title":"Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free Domain Adaptation for Video Semantic Segmentation","paper_url":"/paper/spatio-temporal-pixel-level-contrastive","paper_date":"2023-03-25","arxiv_id":"2303.14361","code_links":[{"title":"shaoyuanlo/stpl","url":"https://github.com/shaoyuanlo/stpl"}],"syntology":null}}],"datasets":[{"url":"/dataset/pacs","name":"PACS","full_name":"Photo-Art-Cartoon-Sketch","num_papers_in_archive":668},{"url":"/dataset/synthia","name":"SYNTHIA","full_name":"SYNTHetic Collection of Imagery and Annotations","num_papers_in_archive":538},{"url":"/dataset/gta5","name":"GTA5","full_name":"Grand Theft Auto 5","num_papers_in_archive":412},{"url":"/dataset/visda-2017","name":"VisDA-2017","full_name":"VisDA-2017","num_papers_in_archive":223},{"url":"/dataset/dark-zurich","name":"Dark Zurich","full_name":"","num_papers_in_archive":57},{"url":"/dataset/acdc-adverse-conditions-dataset-with","name":"ACDC (Adverse Conditions Dataset with Correspondences)","full_name":"Adverse Conditions Dataset with Correspondences","num_papers_in_archive":31}],"subtasks":[{"url":"/task/3d-source-free-domain-adaptation","name":"3D Source-Free Domain Adaptation"},{"url":"/task/source-free-object-detection","name":"Source Free Object Detection"}],"parent_tasks":[{"url":"/task/domain-adaptation","name":"Domain Adaptation"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":103,"tagged_in_all":188,"items":[{"url":"/paper/upcycling-models-under-domain-and-category","title":"Upcycling Models under Domain and Category Shift","date":"2023-03-13","arxiv_id":"2303.07110","repositories_listed":3,"syntology":null},{"url":"/paper/concurrent-subsidiary-supervision-for","title":"Concurrent Subsidiary Supervision for Unsupervised Source-Free Domain Adaptation","date":"2022-07-27","arxiv_id":"2207.13247","repositories_listed":3,"syntology":{"n":13,"n_ran":6,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/do-we-really-need-to-access-the-source-data","title":"Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation","date":"2020-02-20","arxiv_id":"2002.08546","repositories_listed":3,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/leveraging-segment-anything-model-for-source","title":"Leveraging Segment Anything Model for Source-Free Domain Adaptation via Dual Feature Guided Auto-Prompting","date":"2025-05-13","arxiv_id":"2505.08527","repositories_listed":2,"syntology":null},{"url":"/paper/glc-source-free-universal-domain-adaptation","title":"GLC++: Source-Free Universal Domain Adaptation through Global-Local Clustering and Contrastive Affinity Learning","date":"2024-03-21","arxiv_id":"2403.14410","repositories_listed":2,"syntology":null},{"url":"/paper/towards-source-free-domain-adaptive-semantic","title":"Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning","date":"2023-06-02","arxiv_id":"2306.01598","repositories_listed":2,"syntology":null},{"url":"/paper/guiding-pseudo-labels-with-uncertainty","title":"Guiding Pseudo-labels with Uncertainty Estimation for Source-free Unsupervised Domain Adaptation","date":"2023-03-07","arxiv_id":"2303.03770","repositories_listed":2,"syntology":{"n":3,"n_ran":1,"n_unverified":2,"n_pointer_only":3}},{"url":"/paper/divide-and-contrast-source-free-domain","title":"Divide and Contrast: Source-free Domain Adaptation via Adaptive Contrastive Learning","date":"2022-11-12","arxiv_id":"2211.06612","repositories_listed":2,"syntology":{"n":18,"n_ran":7,"n_unverified":11,"n_pointer_only":15}},{"url":"/paper/proxymix-proxy-based-mixup-training-with","title":"ProxyMix: Proxy-based Mixup Training with Label Refinery for Source-Free Domain Adaptation","date":"2022-05-29","arxiv_id":"2205.14566","repositories_listed":2,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/exploiting-the-intrinsic-neighborhood","title":"Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation","date":"2021-10-08","arxiv_id":"2110.04202","repositories_listed":2,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/source-data-absent-unsupervised-domain","title":"Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfer","date":"2020-12-14","arxiv_id":"2012.07297","repositories_listed":2,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/safe-self-attention-based-unsupervised-road","title":"SS-SFDA : Self-Supervised Source-Free Domain Adaptation for Road Segmentation in Hazardous Environments","date":"2020-11-27","arxiv_id":"2012.08939","repositories_listed":2,"syntology":null},{"url":"/paper/unsupervised-domain-adaptation-without-source","title":"Casting a BAIT for Offline and Online Source-free Domain Adaptation","date":"2020-10-23","arxiv_id":"2010.12427","repositories_listed":2,"syntology":null},{"url":"/paper/fully-test-time-adaptation-by-entropy","title":"Tent: Fully Test-time Adaptation by Entropy Minimization","date":"2020-06-18","arxiv_id":"2006.10726","repositories_listed":2,"syntology":{"n":9,"n_ran":8,"n_unverified":1,"n_pointer_only":3}},{"url":"/paper/srpl-sfda-sam-guided-reliable-pseudo-labels","title":"SRPL-SFDA: SAM-Guided Reliable Pseudo-Labels for Source-Free Domain Adaptation in Medical Image Segmentation","date":"2025-06-11","arxiv_id":"2506.09403","repositories_listed":1,"syntology":null},{"url":"/paper/shuffle-patchmix-augmentation-with-confidence","title":"Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation","date":"2025-05-30","arxiv_id":"2505.24216","repositories_listed":1,"syntology":null},{"url":"/paper/ddfp-data-dependent-frequency-prompt-for","title":"DDFP: Data-dependent Frequency Prompt for Source Free Domain Adaptation of Medical Image Segmentation","date":"2025-05-15","arxiv_id":"2505.09927","repositories_listed":1,"syntology":null},{"url":"/paper/effective-dual-region-augmentation-for","title":"Effective Dual-Region Augmentation for Reduced Reliance on Large Amounts of Labeled Data","date":"2025-04-17","arxiv_id":"2504.13077","repositories_listed":1,"syntology":null},{"url":"/paper/disentangled-source-free-personalization-for","title":"Disentangled Source-Free Personalization for Facial Expression Recognition with Neutral Target Data","date":"2025-03-26","arxiv_id":"2503.20771","repositories_listed":1,"syntology":null},{"url":"/paper/source-free-domain-adaptation-based-on-label","title":"Source-free domain adaptation based on label reliability for cross-domain bearing fault diagnosis","date":"2025-03-11","arxiv_id":"2503.08749","repositories_listed":1,"syntology":null},{"url":"/paper/aif-sfda-autonomous-information-filter-driven","title":"AIF-SFDA: Autonomous Information Filter-driven Source-Free Domain Adaptation for Medical Image Segmentation","date":"2025-01-06","arxiv_id":"2501.03074","repositories_listed":1,"syntology":null},{"url":"/paper/bridge-then-begin-anew-generating-target","title":"Bridge then Begin Anew: Generating Target-relevant Intermediate Model for Source-free Visual Emotion Adaptation","date":"2024-12-18","arxiv_id":"2412.13577","repositories_listed":1,"syntology":null},{"url":"/paper/what-has-been-overlooked-in-contrastive","title":"What Has Been Overlooked in Contrastive Source-Free Domain Adaptation: Leveraging Source-Informed Latent Augmentation within Neighborhood Context","date":"2024-12-18","arxiv_id":"2412.14301","repositories_listed":1,"syntology":{"n":4,"n_ran":4,"n_unverified":0,"n_pointer_only":4}},{"url":"/paper/recall-and-refine-a-simple-but-effective","title":"Recall and Refine: A Simple but Effective Source-free Open-set Domain Adaptation Framework","date":"2024-11-19","arxiv_id":"2411.12558","repositories_listed":1,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":2}},{"url":"/paper/gala-graph-diffusion-based-alignment-with","title":"GALA: Graph Diffusion-based Alignment with Jigsaw for Source-free Domain Adaptation","date":"2024-10-22","arxiv_id":"2410.16606","repositories_listed":1,"syntology":null},{"url":"/paper/a3-active-adversarial-alignment-for-source","title":"A3: Active Adversarial Alignment for Source-Free Domain Adaptation","date":"2024-09-27","arxiv_id":"2409.18418","repositories_listed":1,"syntology":null},{"url":"/paper/source-free-domain-adaptation-for-yolo-object","title":"Source-Free Domain Adaptation for YOLO Object Detection","date":"2024-09-25","arxiv_id":"2409.16538","repositories_listed":1,"syntology":null},{"url":"/paper/train-till-you-drop-towards-stable-and-robust","title":"Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation","date":"2024-09-06","arxiv_id":"2409.04409","repositories_listed":1,"syntology":{"n":19,"n_ran":10,"n_unverified":9,"n_pointer_only":19}},{"url":"/paper/dynamic-retraining-updating-mean-teacher-for","title":"Dynamic Retraining-Updating Mean Teacher for Source-Free Object Detection","date":"2024-07-23","arxiv_id":"2407.16497","repositories_listed":1,"syntology":null},{"url":"/paper/eianet-a-novel-domain-adaptation-approach-to","title":"EIANet: A Novel Domain Adaptation Approach to Maximize Class Distinction with Neural Collapse Principles","date":"2024-07-23","arxiv_id":"2407.16189","repositories_listed":1,"syntology":null}],"syntology_records":11,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}