Research

Paper

TESTING February 19, 2026

LATA: Laplacian-Assisted Transductive Adaptation for Conformal Uncertainty in Medical VLMs

Authors

Behzad Bozorgtabar, Dwarikanath Mahapatra, Sudipta Roy, Muzammal Naseer, Imran Razzak, Zongyuan Ge

Abstract

Medical vision-language models (VLMs) are strong zero-shot recognizers for medical imaging, but their reliability under domain shift hinges on calibrated uncertainty with guarantees. Split conformal prediction (SCP) offers finite-sample coverage, yet prediction sets often become large (low efficiency) and class-wise coverage unbalanced-high class-conditioned coverage gap (CCV), especially in few-shot, imbalanced regimes; moreover, naively adapting to calibration labels breaks exchangeability and voids guarantees. We propose \texttt{\textbf{LATA}} (Laplacian-Assisted Transductive Adaptation), a \textit{training- and label-free} refinement that operates on the joint calibration and test pool by smoothing zero-shot probabilities over an image-image k-NN graph using a small number of CCCP mean-field updates, preserving SCP validity via a deterministic transform. We further introduce a \textit{failure-aware} conformal score that plugs into the vision-language uncertainty (ViLU) framework, providing instance-level difficulty and label plausibility to improve prediction set efficiency and class-wise balance at fixed coverage. \texttt{\textbf{LATA}} is black-box (no VLM updates), compute-light (windowed transduction, no backprop), and includes an optional prior knob that can run strictly label-free or, if desired, in a label-informed variant using calibration marginals once. Across \textbf{three} medical VLMs and \textbf{nine} downstream tasks, \texttt{\textbf{LATA}} consistently reduces set size and CCV while matching or tightening target coverage, outperforming prior transductive baselines and narrowing the gap to label-using methods, while using far less compute. Comprehensive ablations and qualitative analyses show that \texttt{\textbf{LATA}} sharpens zero-shot predictions without compromising exchangeability.

Metadata

arXiv ID: 2602.17535
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-02-19
Fetched: 2026-02-21 18:51

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.17535v1</id>\n    <title>LATA: Laplacian-Assisted Transductive Adaptation for Conformal Uncertainty in Medical VLMs</title>\n    <updated>2026-02-19T16:45:38Z</updated>\n    <link href='https://arxiv.org/abs/2602.17535v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.17535v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Medical vision-language models (VLMs) are strong zero-shot recognizers for medical imaging, but their reliability under domain shift hinges on calibrated uncertainty with guarantees. Split conformal prediction (SCP) offers finite-sample coverage, yet prediction sets often become large (low efficiency) and class-wise coverage unbalanced-high class-conditioned coverage gap (CCV), especially in few-shot, imbalanced regimes; moreover, naively adapting to calibration labels breaks exchangeability and voids guarantees. We propose \\texttt{\\textbf{LATA}} (Laplacian-Assisted Transductive Adaptation), a \\textit{training- and label-free} refinement that operates on the joint calibration and test pool by smoothing zero-shot probabilities over an image-image k-NN graph using a small number of CCCP mean-field updates, preserving SCP validity via a deterministic transform. We further introduce a \\textit{failure-aware} conformal score that plugs into the vision-language uncertainty (ViLU) framework, providing instance-level difficulty and label plausibility to improve prediction set efficiency and class-wise balance at fixed coverage. \\texttt{\\textbf{LATA}} is black-box (no VLM updates), compute-light (windowed transduction, no backprop), and includes an optional prior knob that can run strictly label-free or, if desired, in a label-informed variant using calibration marginals once. Across \\textbf{three} medical VLMs and \\textbf{nine} downstream tasks, \\texttt{\\textbf{LATA}} consistently reduces set size and CCV while matching or tightening target coverage, outperforming prior transductive baselines and narrowing the gap to label-using methods, while using far less compute. Comprehensive ablations and qualitative analyses show that \\texttt{\\textbf{LATA}} sharpens zero-shot predictions without compromising exchangeability.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-02-19T16:45:38Z</published>\n    <arxiv:comment>18 pages, 6 figures, 4 tables</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Behzad Bozorgtabar</name>\n    </author>\n    <author>\n      <name>Dwarikanath Mahapatra</name>\n    </author>\n    <author>\n      <name>Sudipta Roy</name>\n    </author>\n    <author>\n      <name>Muzammal Naseer</name>\n    </author>\n    <author>\n      <name>Imran Razzak</name>\n    </author>\n    <author>\n      <name>Zongyuan Ge</name>\n    </author>\n  </entry>"
}