{"ID":2899491,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.00462","arxiv_id":"2507.00462","title":"Unleashing the Potential of All Test Samples: Mean-Shift Guided Test-Time Adaptation","abstract":"Visual-language models (VLMs) like CLIP exhibit strong generalization but struggle with distribution shifts at test time. Existing training-free test-time adaptation (TTA) methods operate strictly within CLIP's original feature space, relying on high-confidence samples while overlooking the potential of low-confidence ones. We propose MS-TTA, a training-free approach that enhances feature representations beyond CLIP's space using a single-step k-nearest neighbors (kNN) Mean-Shift. By refining all test samples, MS-TTA improves feature compactness and class separability, leading to more stable adaptation. Additionally, a cache of refined embeddings further enhances inference by providing Mean Shift enhanced logits. Extensive evaluations on OOD and cross-dataset benchmarks demonstrate that MS-TTA consistently outperforms state-of-the-art training-free TTA methods, achieving robust adaptation without requiring additional training.","short_abstract":"Visual-language models (VLMs) like CLIP exhibit strong generalization but struggle with distribution shifts at test time. Existing training-free test-time adaptation (TTA) methods operate strictly within CLIP's original feature space, relying on high-confidence samples while overlooking the potential of low-confidence...","url_abs":"https://arxiv.org/abs/2507.00462","url_pdf":"https://arxiv.org/pdf/2507.00462v2","authors":"[\"Jizhou Han\",\"Chenhao Ding\",\"SongLin Dong\",\"Yuhang He\",\"Xinyuan Gao\",\"Yihong Gong\"]","published":"2025-07-01T06:22:00Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Language Model\"]","has_code":false}