From Theory to Decision Rule: Calibrating the Noisy-Label Crossover for Vision-Language Model Weak Supervision Across Three Medical-Imaging Benchmarks

• URL: http://arxiv.org/abs/2605.24771v1
• Date: Sat, 23 May 2026 23:15:07 GMT
• Title: From Theory to Decision Rule: Calibrating the Noisy-Label Crossover for Vision-Language Model Weak Supervision Across Three Medical-Imaging Benchmarks
• Abstract summary: We provide a benchmark calibration for foundation-model weak labels.<n>The crossover predicted by theory appears at ng100 on PCAM, 20-50 on ISIC, and 250-500 on NIH-CXR.<n>A structured-vs-random noise sign flip on NIH-CXR shows that the rate-only formulation of the bound is incomplete.
• Score: 41.99844472131922
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Classical noisy-label theory predicts that downstream performance under weak supervision is bounded above by the labeler's accuracy, implying a sharp crossover: once a gold-trained classifier matches the labeler, weak labels stop helping and start hurting. The prediction is theoretical; what is missing is a benchmark calibration that turns it into an instance-level statement for modern foundation-model labelers. We provide such a calibration for BiomedCLIP-generated weak labels on three medical-imaging benchmarks (PCAM, ISIC, NIH-CXR) and six downstream architectures spanning an 11x parameter range. The crossover predicted by theory appears at ng~100 on PCAM, 20-50 on ISIC, and 250-500 on NIH-CXR; weak labels above the crossover degrade AUC by up to -0.10. The location is architecture-invariant for four of five pretrained architectures, and a within-family DenseNet sweep (2.5x parameters, identical pretraining) supports the view that the labeler, not the student, is the dominant constraint. The calibration in turn produces a decision rule operable from 10-20 gold labels: compare gold-only AUC to VLM accuracy on the user's gold set. A structured-vs-random noise sign flip on NIH-CXR shows that the rate-only formulation of the bound is incomplete and identifies a concrete refinement (label-space projection) that future benchmarks can be designed to test.

Related papers

• Bounded Adjustment with Reliability-Guided Embedding for Imbalanced Learning with Noisy Labels [7.047239027940408]
  We introduce BARGE (Bounded Adjustment with Reliability-Guided Embeddings), a single-stage objective combining a bounded, prior-adjusted density-power score with reliability-guided angular geometry.<n>We evaluate it on CIFAR-10, CIFAR-100, and Tiny ImageNet under long-tail and step imbalance, clean labels, and 20% and 40% random incorrect-label replacement.
  arXiv  Detail & Related papers  (2026-09-14T21:41:28Z)
• A Fuzzy Rule-based Neuro-Symbolic Approach for Pipe Severity Prediction in Sewer Networks [0.0]
  This study introduces a modular, fuzzy rule-based neuro-symbolic framework that bridges this gap by decoupling neural perception from symbolic reasoning.<n>For reasoning, a DT, specifically Weka's J48, algorithm is trained on ground-truth CODEs and severity labels, and its paths are converted into 19 fixed IF--THEN rules.<n>Overall, the framework combines competitive class-balanced performance with traceable reasoning from predicted CODE degrees to rule supports and severity evidence.
  arXiv  Detail & Related papers  (2026-07-30T16:33:26Z)
• Multi-Label Test-Time Adaptation with Bayesian Conditional Priors [15.878958356965159]
  Conditional Priors Estimation (BCP) is a gradient-free test-time adaptation method that injects label dependency without tuning the backbone.<n>BCP operates without target annotations by estimating anchor-conditioned priors online from the unlabeled test stream.<n>BCP consistently outperforms strong TTA baselines across standard multi-label benchmarks and multiple CLIP backbones.
  arXiv  Detail & Related papers  (2026-06-11T05:29:00Z)
• CLiGNet: Clinical Label-Interaction Graph Network for Medical Specialty Classification from Clinical Transcriptions [0.0]
  We first document this methodological flaw and establish a leakage free benchmark across 40 medical specialties (4966 records)<n>We then introduce CLiGNet, a neural architecture that combines a Bio ClinicalBERT text encoder with a two layer Graph Convolutional Network operating on a specialty label graph constructed from semantic similarity and ICD 10 chapter priors.<n>We provide comprehensive failure analysis covering pairwise specialty confusions, rare class behaviour, document length effects, and token level Integrated Gradients attribution, offering actionable insights for clinical NLP system deployment.
  arXiv  Detail & Related papers  (2026-03-24T03:30:06Z)
• LATA: Laplacian-Assisted Transductive Adaptation for Conformal Uncertainty in Medical VLMs [61.06744611795341]
  Medical vision-language models (VLMs) are strong zero-shot recognizers for medical imaging.<n>We propose texttttextbfLATA (Laplacian-Assisted Transductive Adaptation), a textittraining- and label-free refinement.<n>texttttextbfLATA sharpens zero-shot predictions without compromising exchangeability.
  arXiv  Detail & Related papers  (2026-02-19T16:45:38Z)
• Classifier Calibration at Scale: An Empirical Study of Model-Agnostic Post-Hoc Methods [0.0]
  We study model-agnostic post-hoc calibration methods to improve probabilistic predictions in supervised binary classification.<n>We benchmark 21 widely used classifiers, including linear models, SVMs, tree ensembles (CatBoost, XGBoost, LightGBM)<n>We find that commonly used calibration procedures, most notably Platt scaling and isotonic regression, can systematically degrade proper scoring performance.
  arXiv  Detail & Related papers  (2026-01-19T18:23:36Z)
• Token-based Decision Criteria Are Suboptimal in In-context Learning [2.2973949268669562]
  In-Context Learning (ICL) typically utilizes classification criteria from output probabilities of manually selected label tokens.<n>We propose Hidden, which renounces token probabilities and uses the nearest centroid on the LM's last hidden states.<n>Our experiments on 6 models and 10 classification datasets indicate that Hidden consistently outperforms current token-based baselines.
  arXiv  Detail & Related papers  (2024-06-24T11:16:26Z)
• Retraining with Predicted Hard Labels Provably Increases Model Accuracy [77.71162068832108]
  Retraining can improve the population accuracy obtained by initially training with the given (noisy) labels.<n>We empirically show that retraining selectively on the samples for which the predicted label matches the given label significantly improves label DP training at no extra privacy cost.
  arXiv  Detail & Related papers  (2024-06-17T04:53:47Z)
• Conformal Predictor for Improving Zero-shot Text Classification Efficiency [37.745518881553416]
  We reduce the average inference time for NLI- and NSP-based models by 25.6% and 22.2% respectively. With a suitable CP for each dataset, we reduce the average inference time for NLI- and NSP-based models by 25.6% and 22.2% respectively.
  arXiv  Detail & Related papers  (2022-10-23T05:19:50Z)
• Rethinking Pseudo Labels for Semi-Supervised Object Detection [84.697097472401]
  We introduce certainty-aware pseudo labels tailored for object detection. We dynamically adjust the thresholds used to generate pseudo labels and reweight loss functions for each category to alleviate the class imbalance problem. Our approach improves supervised baselines by up to 10% AP using only 1-10% labeled data from COCO.
  arXiv  Detail & Related papers  (2021-06-01T01:32:03Z)
• SCRIB: Set-classifier with Class-specific Risk Bounds for Blackbox Models [48.374678491735665]
  We introduce Set-classifier with Class-specific RIsk Bounds (SCRIB) to tackle this problem. SCRIB constructs a set-classifier that controls the class-specific prediction risks with a theoretical guarantee. We validated SCRIB on several medical applications, including sleep staging on electroencephalogram (EEG) data, X-ray COVID image classification, and atrial fibrillation detection based on electrocardiogram (ECG) data.
  arXiv  Detail & Related papers  (2021-03-05T21:06:12Z)
• Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection [85.53263670166304]
  One-stage detector basically formulates object detection as dense classification and localization. Recent trend for one-stage detectors is to introduce an individual prediction branch to estimate the quality of localization. This paper delves into the representations of the above three fundamental elements: quality estimation, classification and localization.
  arXiv  Detail & Related papers  (2020-06-08T07:24:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.