I know what you trained last summer: A survey on stealing machine learning models and defences
Machine-Learning-as-a-Service (MLaaS) has become a widespread paradigm, making
even the most complex Machine Learning models available for clients via, eg, a pay-per …
even the most complex Machine Learning models available for clients via, eg, a pay-per …
The false promise of imitating proprietary llms
An emerging method to cheaply improve a weaker language model is to finetune it on
outputs from a stronger model, such as a proprietary system like ChatGPT (eg, Alpaca, Self …
outputs from a stronger model, such as a proprietary system like ChatGPT (eg, Alpaca, Self …
High accuracy and high fidelity extraction of neural networks
In a model extraction attack, an adversary steals a copy of a remotely deployed machine
learning model, given oracle prediction access. We taxonomize model extraction attacks …
learning model, given oracle prediction access. We taxonomize model extraction attacks …
Data-free model extraction
Current model extraction attacks assume that the adversary has access to a surrogate
dataset with characteristics similar to the proprietary data used to train the victim model. This …
dataset with characteristics similar to the proprietary data used to train the victim model. This …
Towards data-free model stealing in a hard label setting
Abstract Machine learning models deployed as a service (MLaaS) are susceptible to model
stealing attacks, where an adversary attempts to steal the model within a restricted access …
stealing attacks, where an adversary attempts to steal the model within a restricted access …
Entangled watermarks as a defense against model extraction
Machine learning involves expensive data collection and training procedures. Model owners
may be concerned that valuable intellectual property can be leaked if adversaries mount …
may be concerned that valuable intellectual property can be leaked if adversaries mount …
Deepsteal: Advanced model extractions leveraging efficient weight stealing in memories
Recent advancements in Deep Neural Networks (DNNs) have enabled widespread
deployment in multiple security-sensitive domains. The need for resource-intensive training …
deployment in multiple security-sensitive domains. The need for resource-intensive training …
Thieves on sesame street! model extraction of bert-based apis
We study the problem of model extraction in natural language processing, in which an
adversary with only query access to a victim model attempts to reconstruct a local copy of …
adversary with only query access to a victim model attempts to reconstruct a local copy of …
Protecting intellectual property of large language model-based code generation apis via watermarks
The rise of large language model-based code generation (LLCG) has enabled various
commercial services and APIs. Training LLCG models is often expensive and time …
commercial services and APIs. Training LLCG models is often expensive and time …
Prediction poisoning: Towards defenses against dnn model stealing attacks
High-performance Deep Neural Networks (DNNs) are increasingly deployed in many real-
world applications eg, cloud prediction APIs. Recent advances in model functionality …
world applications eg, cloud prediction APIs. Recent advances in model functionality …