{"ID":2877622,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.19830","arxiv_id":"2508.19830","title":"Target-Agnostic Calibration under Distribution Shift with Frequency-Aware Gradient Rectification","abstract":"Real-world model deployments inevitably encounter distribution shifts, rendering the confidence estimates of deep neural networks highly unreliable, posing severe risks in safety-critical applications. Existing methods improve calibration via training-time regularization or post-hoc adjustment, but often rely on access to (or simulation of) target domains, limiting practicality. We propose Frequency-aware Gradient Rectification (FGR), a target-agnostic training framework for robust calibration. From a frequency perspective, FGR applies low-pass filtering to a subset of training images to diminish spurious high-frequency cues and encourage the learning of domain-invariant features. However, the associated information loss can degrade In-Distribution (ID) calibration. To resolve this trade-off, FGR treats ID calibration as a hard constraint and rectifies conflicting parameter updates via geometric projection. This ensures a first-order non-increase in the ID calibration objective without introducing an additional loss-balancing coefficient. Extensive experiments on synthetic, real-world, and semantic shift datasets demonstrate that FGR significantly improves calibration under diverse shifts while preserving ID performance, and it remains compatible with post-hoc calibration methods. Our code is available at https://github.com/YilinZhang107/FGR-Calib.","short_abstract":"Real-world model deployments inevitably encounter distribution shifts, rendering the confidence estimates of deep neural networks highly unreliable, posing severe risks in safety-critical applications. Existing methods improve calibration via training-time regularization or post-hoc adjustment, but often rely on access...","url_abs":"https://arxiv.org/abs/2508.19830","url_pdf":"https://arxiv.org/pdf/2508.19830v2","authors":"[\"Yilin Zhang\",\"Cai Xu\",\"You Wu\",\"Ziyu Guan\",\"Wei Zhao\"]","published":"2025-08-27T12:28:26Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[]","has_code":false,"code_links":[{"ID":610399,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2877622,"paper_url":"https://arxiv.org/abs/2508.19830","paper_title":"Target-Agnostic Calibration under Distribution Shift with Frequency-Aware Gradient Rectification","repo_url":"https://github.com/YilinZhang107/FGR-Calib","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}