Experiments on two multi-expert ulcerative colitis endoscopic-image datasets under two ordinal-noise models show that Ord-NLL is competitive with or superior to strong baselines while reducing mean absolute error, and that Ord-NLL+ often yields further gains.
Abstract
Disease severity is often annotated using a small number of discrete severity levels with an inherent order, yet such labels are subjective and often corrupted by label noise biased toward adjacent levels. Conventional methods for learning with noisy labels typically treat label noise as random class flips, overlooking the ordinal structure of these misannotations. We propose Ordinal Negative Label Learning (Ord-NLL), an extension of negative label learning that explicitly incorporates ordinal structure into negative-label sampling. Ord-NLL constructs an ordinal negative-label distribution that assigns higher sampling probability to levels farther from the observed label, thereby encouraging learning that respects ordinal relationships. The method uses a single-term objective derived theoretically and requires neither the noise rate nor the label-transition matrix, making it practical when prior knowledge about the noise process is unavailable. We further introduce Ord-NLL+, which leverages the confidence estimates produced by Ord-NLL for sample selection and retraining. Experiments on two multi-expert ulcerative colitis (UC) endoscopic-image datasets under two ordinal-noise models show that Ord-NLL is competitive with or superior to strong baselines while reducing mean absolute error, and that Ord-NLL+ often yields further gains. Across controlled ordinal-noise experiments, Ord-NLL and Ord-NLL + consistently outperformed conventional NLL and remained competitive with strong noisy-label baselines. In a representative severe-noise setting, Ord-NLL + achieved 0.715 accuracy, 0.305 MAE, and 0.625 macro-F1, outperforming both conventional NLL and the best competing baseline. These results suggest that explicitly incorporating ordinal structure into negative-label learning is an effective strategy for robust severity estimation under noisy ordinal annotations.
Ordinal regression is widely used in scenarios where labels are discrete yet inherently ordered. In practice, however, ordinal labels are often obtained by discretizing underlying continuous semantics through subjective human judgment, resulting in ambiguous boundaries and annotation noise. Such uncertainty challenges existing methods that rely on fixed supervision targets, which may reinforce biased ordering under subjective annotations. To address this limitation, we propose D3O, a dynamic distribution distillation framework that replaces static supervision with training-driven evolution of ordinal label distributions via self-distillation. Specifically, we introduce a contrastive ordinal-aware label enhancement module that leverages vision-language alignment to recover refined label distributions capturing both inter-class ambiguity and instance-level uncertainty. Furthermore, we design a CDF-based cross-layer interaction distillation mechanism to propagate cumulative ordinal structure across network hierarchy, ensuring consistent ordinal geometry in intermediate representations. Extensive experiments on four general ordinal regression tasks demonstrate that our proposed D3O consistently outperforms existing approaches, particularly under severe class imbalance and noisy supervision. These results highlight the effectiveness of dynamic supervision in learning robust ordinal representations beyond fixed targets. The code will be publicly available.
This paper proposes a novel Partial label-based Self-training framework (PaSta) that leverages partial label learning technique to overcome the limitations of existing methods and designs a partial label-based classification model with two well-crafted loss functions to guide the model learning at both label and representation spaces.
Yujing Liu, Yixin Liu, Yu Zheng et al.· 0 citations
Active learning can reduce labeling cost by selecting informative examples, but the most uncertain examples may also be the hardest to label correctly. This study tests whether uncertainty sampling fails because it acquires more corrupted labels or because errors concentrated in difficult regions are especially harmful. Margin-based uncertainty sampling is compared with random sampling under clean labels, random classification noise (RCN), and bounded difficulty-dependent noise on three public binary tabular datasets. The design uses 100 paired seeds, nine expected noise rates from 0 to 0.30, annotation budgets from 20 to 120, and logistic regression with regularization re-selected by cross-validation at every budget. An exposure-matched RCN control aligns mean final acquired corruption, while a clean-label extension reaches budget 400. Under clean labels, uncertainty sampling improved normalized balanced-accuracy area under the learning curve by 1.09 to 1.77 percentage points on all datasets. Difficulty-dependent noise reduced this advantage more than RCN at six of eight rates on Breast Cancer Wisconsin, but at no tested rate on Banknote Authentication or MAGIC Gamma Telescope. Exposure-matched analyses found no corrected evidence for a universal additional penalty from structured error location. On clean MAGIC data, uncertainty sampling improved balanced accuracy while reducing average precision and true-positive rate at fixed false-positive rates. Thus, uncertainty sampling was label-efficient, but its apparent robustness depended on dataset, budget, noise structure, and evaluation metric.
Gaussian-mixture calculations and a medical diagnosis example illustrate how uncertainty-dependent labeling mechanisms can improve estimation and classification under a fixed labeling budget.
You‐Gan Wang, Jinran Wu, Geoffrey J. McLachlan· 0 citations
LDIBR performs instance-adaptive imputation conditioned on instance features and the binary observation mask, and learns a prior, a reliability-gated correction, and entry-wise fusion weights to produce a normalized imputed distribution.
Xiang-Cheng Sun, Miaogen Ling, Han Qin et al.· 0 citations
This work proposes an integrated learning paradigm that simultaneously enhances feature compactness and improves robustness against label noise and introduces a feature disentanglement mechanism that isolates reliable label-related feature representations from spurious ones introduced by noisy supervision.
Yuzhi Tao, Anhui Tan· Computers, Materials & C...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
