Results 101 to 110 of about 918,391 (278)

Zero-Shot Annotation by Large Language Model with Serial Correction of Mixed Label Corruption for Weakly Supervised Financial News Classification

Information
Multi-label classification of financial news is frequently affected by incomplete and noisy annotations, while obtaining expert-curated labels at scale is prohibitively expensive.
Jianxin Sun, Haichuan Liu
doaj   +1 more source

Analysis of the development of modifying loss functions in label-noise environments [PDF]

ITM Web of Conferences
In recent years, the rapid advance of artificial intelligence has intensified interest in deep learning. However, during the training process of deep learning, the paper found that it is often susceptible to label noise.
Jin Liang
doaj   +1 more source

Clean Ocean Advocate, April 2008

, 2008
The monthly Clean Ocean Advocate newsletter informs citizens about critical ocean pollution issues, important meeting dates, proposed legislation, special events, and other notes of interest.

core   +1 more source

Sertraline Treatment Can Mimic Niemann‐Pick Type C Biomarker Profile: A Diagnostic Pitfall

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Background Oxysterols (cholestane‐3β,5α,6β‐triol and 7‐ketocholesterol) and N‐palmitoyl‐O‐phosphocholineserine (PPCS) are sensitive biomarkers for Niemann‐Pick disease type C (NPC) screening. However, false‐positive results occur, with a biomarker profile suggestive of NPC despite the absence of pathogenic variants in genes involved in NPC or ...
Maria Makrygianni, Cecile Pagan, Antony Citterio‐Quentin, Isabelle Rouvet, Giulia Dingeo, Cyril Hanin, Claudine Laurent‐Levinson, Samia Pichard, Bénédicte Héron, Clément Gourguechon, Daniele Mandia, Jerome Guitton, Carole Lacout, Nicolas Schleinitz, Florient Potier, Olivier Flabeau, Elsa Besse‐Pinot, Foudil Lamari, Magali Pettazzoni, Yann Nadjar   +19 more
wiley   +1 more source

Clean-label backdoor attack on link prediction task

Cybersecurity
Graph Neural Networks (GNNs) have shown excellent performance as a powerful tool on link prediction task. Recent studies have shown that link prediction based on GNNs is vulnerable to backdoor attacks.
Junming Mo, Ming Xu, Xiaogang Xing
doaj   +1 more source

Clinical Impact of NOTCH3 Variant Location After First Stroke in CADASIL

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective Despite its monogenic origin, Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy exhibits marked variability in clinical expression and severity. Variants in the NOTCH3 gene, within epidermal growth factor‐like repeat domains 1–6 or 7–34, are known to influence disease onset, but their impact ...
Léa Aguilhon, Hugues Chabriat, Dominique Hervé, Stéphanie Guey, Sophie Tezenas Du Montcel, Juliette Ortholand   +5 more
wiley   +1 more source

Lactic Acid Bacteria Fermentation as a Tool for Creating Texture in Plant-Based “Clean-Label” Cheeses

Fermentation
The growing demand for clean-label plant-based cheese alternatives underscores the need for products with desirable properties. A key technological challenge is replicating the firmness of traditional cheese without synthetic additives.
Valeria Novikova, Anna Shiker, Anastasia Vostrikova, Egor Parkhomenko, Ilya Demyanenko, Olga Korneeva, Zorigto Namsaraev, Alexey Sazonov   +7 more
doaj   +1 more source

Local Dynamic Clean-Label Backdoor Attack with Image Salience Regions [PDF]

Jisuanji kexue yu tansuo
With the widespread application of deep learning technology, there are more and more backdoor attacks on deep learning models. The research on backdoor attacks is of great significance in revealing security risks in the field of artificial intelligence ...
HONG Wei, GENG Peilin, WANG Hongyu, ZHANG Xueqin, GU Chunhua
doaj   +1 more source

Clean Ocean Advocate, September 2009

, 2009
The monthly Clean Ocean Advocate newsletter informs citizens about critical ocean pollution issues, important meeting dates, proposed legislation, special events, and other notes of interest.

core   +1 more source

Multi-label classification using ensembles of pruned sets [PDF]

, 2008
This paper presents a Pruned Sets method (PS) for multi-label classification. It is centred on the concept of treating sets of labels as single labels. This allows the classification process to inherently take into account correlations between labels. By
Pfahringer, Bernhard, Bernhard Pfahringer, Geoff Holmes, Holmes, Geoffrey, Jesse Read, Read, Jesse   +5 more
core   +1 more source