Research

Paper

TESTING March 10, 2026

Predictive Spectral Calibration for Source-Free Test-Time Regression

Authors

Nguyen Viet Tuan Kiet, Huynh Thanh Trung, Pham Huy Hieu

Abstract

Test-time adaptation (TTA) for image regression has received far less attention than its classification counterpart. Methods designed for classification often depend on classification-specific objectives and decision boundaries, making them difficult to transfer directly to continuous regression targets. Recent progress revisits regression TTA through subspace alignment, showing that simple source-guided alignment can be both practical and effective. Building on this line of work, we propose Predictive Spectral Calibration (PSC), a source-free framework that extends subspace alignment to block spectral matching. Instead of relying on a fixed support subspace alone, PSC jointly aligns target features within the source predictive support and calibrates residual spectral slack in the orthogonal complement. PSC remains simple to implement, model-agnostic, and compatible with off-the-shelf pretrained regressors. Experiments on multiple image regression benchmarks show consistent improvements over strong baselines, with particularly clear gains under severe distribution shifts.

Metadata

arXiv ID: 2603.09338
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-10
Fetched: 2026-03-11 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.09338v1</id>\n    <title>Predictive Spectral Calibration for Source-Free Test-Time Regression</title>\n    <updated>2026-03-10T08:15:44Z</updated>\n    <link href='https://arxiv.org/abs/2603.09338v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.09338v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Test-time adaptation (TTA) for image regression has received far less attention than its classification counterpart. Methods designed for classification often depend on classification-specific objectives and decision boundaries, making them difficult to transfer directly to continuous regression targets. Recent progress revisits regression TTA through subspace alignment, showing that simple source-guided alignment can be both practical and effective. Building on this line of work, we propose Predictive Spectral Calibration (PSC), a source-free framework that extends subspace alignment to block spectral matching. Instead of relying on a fixed support subspace alone, PSC jointly aligns target features within the source predictive support and calibrates residual spectral slack in the orthogonal complement. PSC remains simple to implement, model-agnostic, and compatible with off-the-shelf pretrained regressors. Experiments on multiple image regression benchmarks show consistent improvements over strong baselines, with particularly clear gains under severe distribution shifts.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-10T08:15:44Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Nguyen Viet Tuan Kiet</name>\n    </author>\n    <author>\n      <name>Huynh Thanh Trung</name>\n    </author>\n    <author>\n      <name>Pham Huy Hieu</name>\n    </author>\n  </entry>"
}