Open access

CANNE: CLIP-Based ANNE Selection for Noisy-Label Learning

Ge Jin Qian Zhang Li Huang Yongqi Yuan Yu-Long Qiao De Yu Mengjie Ye Ying-Wen Zhu

Jul 2026 · Entropy · Vol 28 · 0 citations · 53 references

Medicine

TL;DR

Experimental results on CIFAR-10, CIFAR-100, Animal-10N, and Mini-WebVision, together with additional evaluation under open-set noise, show that the proposed CANNE method achieves competitive performance across diverse noisy-label settings.

Abstract

Learning with noisy labels (LNL) remains challenging, especially when the identification of clean samples relies heavily on the predictions of the model being trained. In such cases, early-stage selection errors may be reinforced during iterative optimization, leading to unreliable supervision. To alleviate this issue, a two-stage framework, termed CANNE, is proposed by combining Contrastive Language–Image Pre-training (CLIP)-based conservative offline cleaning with Adaptive Nearest Neighbors and eigenvector-based sample selection (ANNE)-based online refinement. Specifically, a high-confidence clean seed set is first constructed using two complementary probability sources derived from frozen CLIP representations and reliability criteria, including class-wise loss modeling and prediction consistency. This seed set is then used as a set of reliable anchors during the subsequent ANNE training process, where online feature- and neighborhood-based refinement further recovers and adjusts sample partitions. In this way, CANNE uses external vision–language priors to provide conservative and persistent guidance while preserving the adaptive recovery ability of online noisy-label learning. Experimental results on CIFAR-10, CIFAR-100, Animal-10N, and Mini-WebVision, together with additional evaluation under open-set noise, show that the proposed method achieves competitive performance across diverse noisy-label settings. In particular, CANNE achieves 96.6% and 96.3% best accuracies on CIFAR-10 under 80% and 90% symmetric noise, respectively, and 81.0% and 79.0% on CIFAR-100 under 20% and 50% symmetric noise. Additional repeated-run, threshold-sensitivity, and runtime analyses further indicate that the CLIP-based seed set provides stable guidance with only moderate computational overhead.

DOI PubMed Semantic Scholar

Read PDF

Similar papers

Aug 2026

Certainty-Aware Partition and Sufficient Utilization of Noisy Samples

CAPSUN is proposed, a robust framework that improves the precision of clean sample selection and mitigates distribution bias through alignment among subsets, and designs a distribution alignment module to adjust the class distribution contrast of labeled and unlabeled subsets to mitigate class distribution discrepancies.

Jia Zhang, Gao-Xia Jiang, Senyu Hou et al. · Machine-mediated learning · 0 citations

Preprint Aug 2026

Triple Expert Learning from Noisy Labels for Semi-Supervised Vision Foundation Model Adaptation

By separating different pseudo-label reliability regions into specialized adaptation paths, TriNoL improves robustness to noisy supervision while keeping the training cost low.

Xuanyu Liu, Zheng Fang, Hong-Yang He et al. · 1 citation

Open access Aug 2026

Semisupervised Adaptation of Vision-Language Models for Image Classification

Results on the UC Merced (UCM) and NWPU benchmarks indicate that SE-CLIP significantly outperforms existing semi-supervised approaches and provides a viable solution for adapting VLMs to the remote sensing domain with minimal human intervention.

M. L. Mekhalfi, M. M. Al Rahhal, Y. Bazi et al. · IEEE Geoscience and Remote S... · 0 citations

Preprint Aug 2026

PaSta: Noisy Node Classification with Partial Label Learning

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

Rethinking BCE Loss for Multi-Label Image Recognition with Fine-Tuning

Class-wise Covariance Regularization is proposed, which aligns the predicted covariance structure of class confidences with the semantic correlations encoded in pretrained text embed-dings with the geometric consistency of the class space throughout fine-tuning, resulting in more stable and interpretable confidence distributions across categories.

Ao Zhou, Zhiwei Jiang, Zifeng Cheng et al. · 0 citations

2026

Audio Active Learning With Noisy Labels

Audio annotation is particularly costly and prone to errors due to the temporal nature and semantic ambiguity in audio perception. Active learning (AL) addresses this by iteratively selecting the most informative samples from an unlabeled pool for expert labeling, thereby maximizing model performance with minimal annotation effort. In the standard continual fine-tuning framework widely adopted in audio AL, the model trained in each cycle serves as the initialization for the next, preserving accumulated knowledge to enhance both sample selection and model accuracy. However, we discover a critical limitation when annotation noise is inevitably introduced: this default continual fine-tuning approach becomes susceptible to Primacy Bias — a phenomenon where early-learned patterns persistently influence subsequent learning. Our experiments show that this bias causes the model to overfit noise more rapidly in later cycles. To our knowledge, this represents the first comprehensive study specifically addressing Active Learning with Noisy Labels (ALNL) in the audio domain. To address this issue, we introduce a re-initialization strategy for ALNL scenarios. Our experiments demonstrate that periodically resetting model parameters preserves the model’s ability to learn from clean samples. Furthermore, we propose the Self-Purify Active Learning (SPAL) method, which dynamically identifies potential label noise via training loss modeling and supports either human-in-the-loop correction or automated label refurbishment. Extensive experiments on underwater acoustics, general audio, and speech datasets demonstrate the effectiveness of our framework against label noise in AL scenarios.

Yi Su, Hui Geng, Qisheng Xu et al. · IEEE Transactions on Audio,... · 0 citations