Yanhui Gu

ExpertFusion: Calibrated multi-expert decision fusion for phishing URL detection under distribution shift and target-prior uncertainty

Musarat Hussain, Junaid Abbas, Jamil Hussain, Yanhui Gu et autres

kr, cn (code pays fourni par la source)

Controlling Prediction Dynamics for Reliable Intraoperative Segmentation

JiuTao Zhou, Xiaoyang Li, Yuhao Zhang, Xiaoqian Peng et autres

Automatic detection and segmentation in intraoperative imaging sequences remains challenging because procedural events can change image appearance abruptly. Instrument motion, material injection, irrigation and suction, and acquisition changes introduce strong artifacts. As a result, models that process each image independently can achieve …

cn (code pays fourni par la source)

Matching the Coupling of Valence Electrons in the Oxide Interface to Perturb the Magnetic Order Enhancing Oxygen Reduction in Zinc–Air Batteries

J F Li, Ningkang Peng, Jianhua Ma, Tingyu Lu et autres

ABSTRACT The inherently locked spin state between the metal sites and oxygen‐containing intermediates imposes an intrinsic limitation on the maximum achievable oxygen reduction reaction (ORR) activity. Herein, we construct the sub‐5 nm Fe 2 O 3 /Sm 2 O 3 heterojunctions immobilized …

cn, jp (code pays fourni par la source)

Matching the Coupling of Valence Electrons in the Oxide Interface to Perturb the Magnetic Order Enhancing Oxygen Reduction in Zinc–Air Batteries

J F Li, Ningkang Peng, Jianhua Ma, Tingyu Lu et autres

ABSTRACT The inherently locked spin state between the metal sites and oxygen‐containing intermediates imposes an intrinsic limitation on the maximum achievable oxygen reduction reaction (ORR) activity. Herein, we construct the sub‐5 nm Fe 2 O 3 /Sm 2 O 3 heterojunctions immobilized …

cn, jp (code pays fourni par la source)

Early High-Frequency Injection for Geometry-Sensitive OOD Detection

Chuanjie Cheng, Ningkang Peng, Chenxi Liu, Yifan He et autres

Post-hoc OOD detectors score logits or features after training, so their success depends on the geometry already encoded in the representation. We revisit this assumption through a band-wise MMD^2 analysis across CE, SimCLR, SupCon, and the OOD-oriented representation method PALM. In our …

Holistic Reliability Propagation: Decoupling Annotation and Prediction for Robust Noisy-Label

Jingyang Mao, Ningkang Peng, Yanhui Gu

Learning with noisy labels in multimedia classification often combines external annotations and model predictions into a single reliability weight, even though the two sources can fail for different reasons. We instead estimate disentangled reliabilities: bilevel meta-learning produces two batch-normalized scalars per sample, …

GAMR: Geometric-Aware Manifold Regularization with Virtual Outlier Synthesis for Learning with Noisy Labels

Ningkang Peng, Jingyang Mao, Xiaoqian Peng, Peirong Ma et autres

Deep neural networks (DNNs) experience significant performance degradation when processing noisy labels, primarily due to overfitting on mislabeled data. Current mainstream approaches attempt to mitigate this issue by passively filtering clean samples during training. However, simple sample filtering within feature spaces degraded …

GAMR: Geometric-Aware Manifold Regularization with Virtual Outlier Synthesis for Learning with Noisy Labels

Ningkang Peng, Jingyang Mao, Xiaoqian Peng, Peirong Ma et autres

Deep neural networks (DNNs) experience significant performance degradation when processing noisy labels, primarily due to overfitting on mislabeled data. Current mainstream approaches attempt to mitigate this issue by passively filtering clean samples during training. However, simple sample filtering within feature spaces degraded …

cn (code pays fourni par la source)

Early High-Frequency Injection for Geometry-Sensitive OOD Detection

Chuanjie Cheng, Ningkang Peng, Chenxi Liu, Yifan He et autres

Post-hoc OOD detectors score logits or features after training, so their success depends on the geometry already encoded in the representation. We revisit this assumption through a band-wise MMD^2 analysis across CE, SimCLR, SupCon, and the OOD-oriented representation method PALM. In our …

cn (code pays fourni par la source)

Holistic Reliability Propagation: Decoupling Annotation and Prediction for Robust Noisy-Label

Jingyang Mao, Ningkang Peng, Yanhui Gu

Learning with noisy labels in multimedia classification often combines external annotations and model predictions into a single reliability weight, even though the two sources can fail for different reasons. We instead estimate disentangled reliabilities: bilevel meta-learning produces two batch-normalized scalars per sample, …

cn (code pays fourni par la source)

Is Complex Training Necessary for Long-Tailed OOD Detection? A Re-think from Feature Geometry

Ningkang Peng, Xuanming Chen, Yanhui Gu

Long-tailed out-of-distribution (LT-OOD) detection is often addressed with specialized training, including auxiliary out-of-distribution (OOD) data, abstention heads, contrastive objectives, energy losses, or gradient-conflict control. We show that these training mechanisms can obscure a simpler issue: frozen long-tailed representations may already contain useful …

When Accuracy Is Not Enough: Uncertainty Collapse between Noisy Label Learning and Out-of-Distribution Detection

Ningkang Peng, Jingyang Mao, Runhan Zhou, Peirong Ma et autres

Learning with noisy labels (LNL) is typically benchmarked by closed-set classification accuracy, yet deployment often requires classifiers to reject out-of-distribution (OOD) inputs. We present a learner-agnostic ACC-OOD benchmark that freezes LNL checkpoints and evaluates them with standardized near-/far-OOD routing and post-hoc scores …