Membership Inference Attacks (MIAs) on pretrained Large Language Models (LLMs) aim at determining if a data point was part of the model's training set.Prior MIAs that are built for classification models fail at LLMs, due to ignoring the generative nature of LLMs across token sequences.In this paper, we present a novel attack on pre-trained LLMs that adapts MIA statistical tests to the perplexity dynamics of subsequences within a data point.Our method significantly outperforms prior approaches, revealing context-dependent memorization patterns in pre-trained LLMs.

Transformers have recently revolutionized the machine learning (ML) landscape, gradually making their way into everyday tasks and equipping our computers with "sparks of intelligence". However, their runtime requirements have prevented them from being broadly deployed on mobile. As personal devices become increasingly powerful at the consumer edge and prompt privacy becomes an ever more pressing issue, we explore the current state of mobile execution of Large Language Models (LLMs). To achieve this, we have created our own automation infrastructure, MELT, which supports the headless execution and benchmarking of LLMs on device, supporting different models, devices and frameworks, including Android, iOS and Nvidia Jetson devices. We evaluate popular instruction fine-tuned LLMs and leverage different frameworks to measure their end-to-end and granular performance, tracing their memory and energy requirements along the way.

Federated Learning (FL) [2] has emerged as a popular solution of Confidential Computing [3] to distributedly train a model on user devices, improving privacy and system scalability. Such privacy-preserving models can be used in wide range of applications, and especially in Telco networks [4]. However, there are no practical systems to easily enable FL training on mobile apps, and especially in an as-a-service fashion. In this demo, we implement and test FLaaS, our recently proposed end-to-end FL service [1]. FLaaS includes a client-side framework with app library and service, and a back-end server, to enable secure and easy to deploy intra- and inter-app FL model training on mobile environments.

Since the inception of Recommender Systems (RS), the accuracy of the recommendations in terms of relevance has been the golden criterion for evaluating the quality of RS algorithms. However, by focusing on item relevance, one pays a significant price in terms of other important metrics: users get stuck in a "filter bubble" and their array of options is significantly reduced, hence degrading the quality of the user experience and leading to churn. Recommendation, and in particular session-based/sequential recommendation, is a complex task with multiple - and often conflicting objectives - that existing state-of-the-art approaches fail to address. In this work, we take on the aforementioned challenge and introduce Scalarized Multi-Objective Reinforcement Learning (SMORL) for the RS setting, a novel Reinforcement Learning (RL) framework that can effectively address multi-objective recommendation tasks. The proposed SMORL agent augments standard recommendation models with additional RL layers that enforce it to simultaneously satisfy three principal objectives: accuracy, diversity, and novelty of recommendations. We integrate this framework with four state-of-the-art session-based recommendation models and compare it with a single-objective RL agent that only focuses on accuracy. Our experimental results on two real-world datasets reveal a substantial increase in aggregate diversity, a moderate increase in accuracy, reduced repetitiveness of recommendations, and demonstrate the importance of reinforcing diversity and novelty as complementary objectives.

We propose and implement a Privacy-preserving Federated Learning ( PPFL ) framework for mobile systems to limit privacy leakages in federated learning. Leveraging the widespread presence of Trusted Execution Environments (TEEs) in high-end and mobile devices, we utilize TEEs on clients for local training, and on servers for secure aggregation, so that model/gradient updates are hidden from adversaries. Challenged by the limited memory size of current TEEs, we leverage greedy layer-wise training to train each model's layer inside the trusted area until its convergence. The performance evaluation of our implementation shows that PPFL can significantly improve privacy while incurring small system overheads at the client-side. In particular, PPFL can successfully defend the trained model against data reconstruction, property inference, and membership inference attacks. Furthermore, it can achieve comparable model utility with fewer communication rounds (0.54×) and a similar amount of network traffic (1.002×) compared to the standard federated learning of a complete model. This is achieved while only introducing up to ~15% CPU time, ~18% memory usage, and ~21% energy consumption overhead in PPFL's client-side.

We present DarkneTZ, a framework that uses an edge device's Trusted Execution Environment (TEE) in conjunction with model partitioning to limit the attack surface against Deep Neural Networks (DNNs). Increasingly, edge devices (smartphones and consumer IoT devices) are equipped with pre-trained DNNs for a variety of applications. This trend comes with privacy risks as models can leak information about their training data through effective membership inference attacks (MIAs).

Internet-of-Things (IoT) devices and applications are being deployed in our homes and workplaces. These devices often rely on continuous data collection to feed machine learning models. However, this approach introduces several privacy and efficiency challenges, as the service operator can perform unwanted inferences on the available data. Recently, advances in edge processing have paved the way for more efficient, and private, data processing at the source for simple tasks and lighter models, though they remain a challenge for larger and more complicated models. In this article, we present a hybrid approach for breaking down large, complex deep neural networks for cooperative, and privacy-preserving analytics. To this end, instead of performing the whole operation on the cloud, we let an IoT device to run the initial layers of the neural network, and then send the output to the cloud to feed the remaining layers and produce the final result. In order to ensure that the user's device contains no extra information except what is necessary for the main task and preventing any secondary inference on the data, we introduce Siamese fine-tuning. We evaluate the privacy benefits of this approach based on the information exposed to the cloud service. We also assess the local inference cost of different layers on a modern handset. Our evaluations show that by using Siamese fine-tuning and at a small processing cost, we can greatly reduce the level of unnecessary, potentially sensitive information in the personal data, thus achieving the desired tradeoff between utility, privacy, and performance.

We present and evaluate Deep Private-Feature Extractor (DPFE), a deep model which is trained and evaluated based on information theoretic constraints. Using the selective exchange of information between a user's device and a service provider, DPFE enables the user to prevent certain sensitive information from being shared with a service provider, while allowing them to extract approved information using their model. We introduce and utilize the log-rank privacy, a novel measure to assess the effectiveness of DPFE in removing sensitive information and compare different models based on their accuracy-privacy trade-off. We then implement and evaluate the performance of DPFEon smartphones to understand its complexity, resource demands, and efficiency trade-offs. Our results on benchmark image datasets demonstrate that under moderate resource utilization, DPFE can achieve high accuracy for primary tasks while preserving the privacy of sensitive information.

Pre-trained Deep Neural Network (DNN) models are increasingly used in smartphones and other user devices to enable prediction services, leading to potential disclosures of (sensitive) information from training data captured inside these models. Based on the concept of generalization error, we propose a framework to measure the amount of sensitive information memorized in each layer of a DNN. Our results show that, when considered individually, the last layers encode a larger amount of information from the training data compared to the first layers. We find that the same DNN architecture trained with different datasets has similar exposure per layer. We evaluate an architecture to protect the most sensitive layers within an on-device Trusted Execution Environment (TEE) against potential white-box membership inference attacks without the significant computational overhead.

Recent advances in cloud computing have simplified the way that both software development and testing are performed. Unfortunately, this is not true for battery testing for which state of the art test-beds simply consist of one phone attached to a power meter. These test-beds have limited resources, access, and are overall hard to maintain; for these reasons, they often sit idle with no experiment to run. In this paper, we propose to share existing battery testing setups and build BatteryLab, a distributed platform for battery measurements. Our vision is to transform independent battery testing setups into vantage points of a planetary-scale measurement platform offering heterogeneous devices and testing conditions. In the paper, we design and deploy a combination of hardware and software solutions to enable BatteryLab's vision. We then preliminarily evaluate BatteryLab's accuracy of battery reporting, along with some system benchmarking. We also demonstrate how BatteryLab can be used by researchers to investigate a simple research question.

There has been a growing interest in measuring and optimizing the power efficiency of mobile apps. Traditional power evaluations rely either on inaccurate software-based solutions or on ad-hoc testbeds composed of a power meter and a mobile device. This demonstration presents BatteryLab, our solution to share existing battery testing setups to build a distributed platform for battery measurements. Our vision is to transform independent battery testing setups into vantage points of a planetary-scale measurement platform offering heterogeneous devices and testing conditions. We demonstrate BatteryLab functionalities by investigating the energy efficiency of popular websites when loaded via both Android and iOS browsers. Our demonstration is also live at https://batterylab.dev/.

Not all smartphone owners use their device in the same way. In this work, we uncover broad, latent patterns of mobile phone use behavior. We conducted a study where, via a dedicated logging app, we collected daily mobile phone activity data from a sample of 340 participants for a period of four weeks. Through an unsupervised learning approach and a methodologically rigorous analysis, we reveal five generic phone use profiles which describe at least 10% of the participants each: limited use, business use, power use, and personality- & externally induced problematic use. We provide evidence that intense mobile phone use alone does not predict negative well-being. Instead, our approach automatically revealed two groups with tendencies for lower well-being, which are characterized by nightly phone use sessions.

Researchers from different disciplines have examined crowd behavior in the past by employing a variety of methods including ethnographic studies, computer vision techniques and manual annotation based data analysis. However, because of the inherent difficulties in collecting, processing and analyzing the data, it is difficult to obtain large data sets for study. In this work we present a system for detecting stationary interactions inside crowds, depending entirely on the sensors available in a modern smartphone device such as Bluetooth Smart (BLE) and Accelerometer. By utilizing Apple's iBeaconTM implementation of Bluetooth Smart using SensingKit1, our open-source multi-platform mobile sensing framework [1], we are able to detect the proximity of users carrying a smartphone in their pocket. We then use an algorithm based on graph theory to predict group interactions inside the crowd. Previous work in this area has been limited to the detection of interactions between only two people and therefore our approach goes beyond current state of the art in its ability to detect group formations with more than two people involved. Our approach is particularly beneficial to the design and implementation of crowd behavior analytics, design of influence strategies, and algorithms for crowd reconfiguration.

Today's smartphones come equipped with a range of advanced sensors capable of sensing motion, orientation, audio as well as environmental data with high accuracy. With the existence of application distribution channels such as the Apple App Store and the Google Play Store, researchers can distribute applications and collect large scale data in ways that previously were not possible. Motivated by the lack of a universal, multi-platform sensing library, in this work we present the design and implementation of SensingKit, an open-source continuous sensing system that supports both iOS and Android mobile devices. One of the unique features of SensingKit is the support of the latest beacon technologies based on Bluetooth Smart (BLE), such as iBeacon and Eddystone. We evaluate and compare the power consumption of each supported sensor individually, using an iPhone 5S device running on iOS 9. We believe that this platform will be beneficial to all researchers and developers who plan to use mobile sensing technology in large-scale experiments.

With the rapid rise in variety of available smartphones today and their rich sensing capabilities, there is an increasing interest in using mobile sensing in large-scale experiments and commercial applications. Motivated by the lack of a universal, multi-platform library, in this paper we present SensingKit, an efficient, open-source, client-server system that supports both iOS and Android mobile devices. SensingKit is capable of continuous sensing the device's motion (Accelerometer, Gyroscope, Magnetometer), location (GPS) and proximity to other smartphones (Bluetooth Smart). The data are temporarily saved to the device's memory and transmitted to a server for further analysis over any Internet connection. We believe that this platform will be beneficial to all researchers and developers who need to perform mobile sensing in their applications and experiments.