Results 11 to 20 of about 835,873 (364)
Weight Poisoning Attacks on Pretrained Models [PDF]
Annual Meeting of the Association for Computational Linguistics, 2020 Recently, NLP has seen a surge in the usage of large pre-trained models. Users download weights of models pre-trained on large datasets, then fine-tune the weights on a task of their choice.
Keita Kurita, Paul Michel, Graham Neubig
semanticscholar +1 more source
Effects of commercial beverages on the neurobehavioral motility of Caenorhabditis elegans [PDF]
PeerJ, 2022 To study the effects of different types of commercially available drinks/beverages on neurobehavior using the model organism C. elegans, and critically review their potential health hazards.
Wenjing Zhang +10 more
doaj +2 more sources
Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoning [PDF]
Conference on Empirical Methods in Natural Language Processing, 2021 Pre-Trained Models have been widely applied and recently proved vulnerable under backdoor attacks: the released pre-trained weights can be maliciously poisoned with certain triggers. When the triggers are activated, even the fine-tuned model will predict
Linyang Li +5 more
semanticscholar +1 more source
Неотложная медицинская помощь, 2023 The review is devoted to the role of psychological factors in the recovery of patients after spinal surgery (with spinal canal stenosis). The high epidemiological and economic significance of the back pain syndrome was noted.
A. Yu. Suroyegina +3 more
doaj +1 more source
Advanced Science, 2023 Cadmium (Cd) is a high‐risk pathogenic toxin for hepatic diseases. Excessive mitophagy is a hallmark in Cd‐induced hepatotoxicity. However, the underlying mechanism remains obscure.
Cong Liu +9 more
doaj +1 more source
Property Inference from Poisoning [PDF]
IEEE Symposium on Security and Privacy, 2021 Property inference attacks consider an adversary who has access to a trained ML model and tries to extract some global statistics of the training data. In this work, we study property inference in scenarios where the adversary can maliciously control a ...
Melissa Chase +2 more
semanticscholar +1 more source
De-Pois: An Attack-Agnostic Defense against Data Poisoning Attacks [PDF]
IEEE Transactions on Information Forensics and Security, 2021 Machine learning techniques have been widely applied to various applications. However, they are potentially vulnerable to data poisoning attacks, where sophisticated attackers can disrupt the learning procedure by injecting a fraction of malicious ...
Jian Chen +4 more
semanticscholar +1 more source
Photochemical Dermatitis Due to Contact With Sosnovsky Hogweed
Неотложная медицинская помощь, 2021 Introduction. Today, the spread of the Sosnovsky hogweed plant has acquired an unprecedented scale, which is accompanied by an annual increase in the number of victims due to the occurrence of photochemical dermatitis (PD).Aim of study.
A. Yu. Simonova +6 more
doaj +1 more source
Неотложная медицинская помощь, 2021 Background. In acute poisoning, accompanied by a violation of the parameters of homeostasis, the problem of its management by the enteral route has been insufficiently studied.Purpose of the study.
V. A. Matkevich +5 more
doaj +1 more source
Aluminum Poisoning with Emphasis on Its Mechanism and Treatment of Intoxication
Emergency Medicine International, 2022 Aluminum poisoning has been reported in some parts of the world. It is one of the global health problems that affect many organs. Aluminum is widely used daily by humans and industries. Residues of aluminum compounds can be found in drinking water, food,
M. R. Rahimzadeh +5 more
semanticscholar +1 more source