Data valuation, a critical aspect of data-centric ML research, aims to
q...
Current literature, aiming to surpass the "Chain-of-Thought" approach, o...
Data valuation – quantifying the contribution of individual data sources...
Data-free knowledge distillation (KD) helps transfer knowledge from a
pr...
Traditionally, data valuation is posed as a problem of equitably splitti...
Data valuation is a growing research field that studies the influence of...
The Shapley value (SV) has emerged as a promising method for data valuat...
Backdoor data detection is traditionally studied in an end-to-end superv...
We study the expressibility and learnability of convex optimization solu...
Data valuation, especially quantifying data value in algorithmic predict...
Data valuation is an essential task in a data marketplace. It aims at fa...
Given the volume of data needed to train modern machine learning models,...
Previous works have validated that text generation APIs can be stolen th...
It is becoming increasingly common to utilize pre-trained models provide...
This paper studies the robustness of data valuation to noisy model
perfo...
With the increasing adoption of NLP models in real-world products, it be...
Backdoor attacks insert malicious data into a training set so that, duri...
Recent studies show that the state-of-the-art deep neural networks are
v...
Machine learning (ML) models need to be frequently retrained on changing...
We propose a minimax formulation for removing backdoors from a given poi...
With the increasing adoption of language models in applications involvin...
Active learning (AL) aims at reducing labeling effort by identifying the...
For most machine learning (ML) tasks, evaluating learning performance on...
High-quality data is critical to train performant Machine Learning (ML)
...
Active learning has been a main solution for reducing data labeling cost...
Backdoor attacks have been considered a severe security threat to deep
l...
Deep learning techniques have achieved remarkable performance in wide-ra...
Model inversion (MI) attacks in the whitebox setting are aimed at
recons...
Large-scale language models such as BERT have achieved state-of-the-art
...
Federated learning (FL) is a popular technique to train machine learning...
This paper studies defense mechanisms against model inversion (MI) attac...
Ever since its proposal, differential privacy has become the golden stan...
Deep neural networks (DNNs) have achieved tremendous success in various
...
This paper studies model-inversion attacks, in which the access to a mod...
This paper focuses on valuating training data for supervised learning ta...
Outlier detection and novelty detection are two important topics for ano...
Given a data set D containing millions of data points and a data
consume...
"How much is my data worth?" is an increasingly common question posed by...
Despite the great success achieved in machine learning (ML), adversarial...
This paper proposes a fundamental answer to a frequently asked question ...
We present results from a set of experiments in this pilot study to
inve...