Search Results for author:
Found 17 papers, 9 papers with code
A Sample-Level Evaluation and Generative Framework for Model Inversion Attacks
Model Inversion (MI) attacks, which reconstruct the training dataset of neural networks, pose significant privacy concerns in machine learning.
FUNU: Boosting Machine Unlearning Efficiency by Filtering Unnecessary Unlearning
To decrease the execution time of such machine unlearning methods, we aim to reduce the size of data removal requests based on the fundamental assumption that the removal of certain data would not result in a distinguishable retrained model.
RALAD: Bridging the Real-to-Sim Domain Gap in Autonomous Driving with Retrieval-Augmented Learning
RALAD features three primary designs, including (1) domain adaptation via an enhanced Optimal Transport (OT) method that accounts for both individual and grouped image distances, (2) a simple and unified framework that can be applied to various models, and (3) efficient fine-tuning techniques that freeze the computationally expensive layers while maintaining robustness.
Fine-tuning is Not Fine: Mitigating Backdoor Attacks in GNNs with Limited Clean Data
By adopting the graph attention transfer method, GRAPHNAD can effectively align the intermediate-layer attention representations of the backdoored model with that of the teacher model, forcing the backdoor neurons to transform into benign ones.
Structure-Preference Enabled Graph Embedding Generation under Differential Privacy
Existing methods tackle this issue by developing deep graph learning models with differential privacy (DP).
Don't Lose Yourself: Boosting Multimodal Recommendation via Reducing Node-neighbor Discrepancy in Graph Convolutional Network
The rapid expansion of multimedia contents has led to the emergence of multimodal recommendation systems.
ModeSeq: Taming Sparse Multimodal Motion Prediction with Sequential Mode Modeling
Anticipating the multimodality of future events lays the foundation for safe autonomous driving.
New Paradigm of Adversarial Training: Breaking Inherent Trade-Off between Accuracy and Robustness via Dummy Classes
Accordingly, to relax the tension between clean and robust learning derived from this overstrict assumption, we propose a new AT paradigm by introducing an additional dummy class for each original class, aiming to accommodate the hard adversarial samples with shifted distribution after perturbation.
"Yes, My LoRD." Guiding Language Model Extraction with Locality Reinforced Distillation
Model extraction attacks (MEAs) on large language models (LLMs) have received increasing attention in recent research.
Why Are My Prompts Leaked? Unraveling Prompt Extraction Threats in Customized Large Language Models
In this paper, we analyze the underlying mechanism of prompt leakage, which we refer to as prompt memorization, and develop corresponding defending strategies.
Understanding is Compression
The better a large model understands the data, the better LMCompress compresses.
Ranking LLMs by compression
We conceptualize the process of understanding as information compression, and propose a method for ranking large language models (LLMs) based on lossless data compression.
BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction
Simulating realistic behaviors of traffic agents is pivotal for efficiently validating the safety of autonomous driving systems.
RSTAR4D: Rotational Streak Artifact Reduction in 4D CBCT using a Separable 4D CNN
We find that streak artifacts exhibit a unique rotational motion along with the patient's respiration, distinguishable from diaphragm-driven respiratory motion in the spatiotemporal domain.
DPSUR: Accelerating Differentially Private Stochastic Gradient Descent Using Selective Update and Release
Motivated by this, this paper proposes DPSUR, a Differentially Private training framework based on Selective Updates and Release, where the gradient from each iteration is evaluated based on a validation test, and only those updates leading to convergence are applied to the model.
TSFool: Crafting Highly-Imperceptible Adversarial Time Series through Multi-Objective Attack
Recent years have witnessed the success of recurrent neural network (RNN) models in time series classification (TSC).
Exploiting Pre-Trained ASR Models for Alzheimer's Disease Recognition Through Spontaneous Speech
Input to these classifiers are speech transcripts produced by automatic speech recognition (ASR) models.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
