SOTAVerified

Data Poisoning

Data Poisoning is an adversarial attack that tries to manipulate the training dataset in order to control the prediction behavior of a trained model such that the model will label malicious examples into a desired classes (e.g., labeling spam e-mails as safe).

Source: Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 81–90 of 492 papers

TitleStatusHype
A Bayesian Incentive Mechanism for Poison-Resilient Federated Learning—0
Self-Adaptive and Robust Federated Spectrum Sensing without Benign Majority for Cellular Networks—0
Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs—0
Addressing The Devastating Effects Of Single-Task Data Poisoning In Exemplar-Free Continual LearningCode0
Tuning without Peeking: Provable Privacy and Generalization Bounds for LLM Post-Training—0
Generalization under Byzantine & Poisoning Attacks: Tight Stability Bounds in Robust Distributed Learning—0
Winter Soldier: Backdooring Language Models at Pre-Training with Indirect Data Poisoning—0
TED-LaST: Towards Robust Backdoor Defense Against Adaptive Attacks—0
Data Shifts Hurt CoT: A Theoretical Study—0
Devil's Hand: Data Poisoning Attacks to Locally Private Graph Learning Protocols—0
Show:102550
← PrevPage 9 of 50Next →

No leaderboard results yet.