Related papers: Don't Lie to Me: Avoiding Malicious Explanations w…

Chaos-based approaches are frequently proposed in information hiding, but without obvious justification. Indeed, the reason why chaos is useful to tackle with discretion, robustness, or security, is rarely elucidated. This research work…

As large language models (LLMs) grow more capable, concerns about their safe deployment have also grown. Although alignment mechanisms have been introduced to deter misuse, they remain vulnerable to carefully designed adversarial prompts.…

Computation offloading (often to external computing resources over a network) has become a necessity for modern applications. At the same time, the proliferation of machine learning techniques has empowered malicious actors to use such…

Many commonly used learning algorithms work by iteratively updating an intermediate solution using one or a few data points in each iteration. Analysis of differential privacy for such algorithms often involves ensuring privacy of each step…

The advancements of Large Language Models (LLMs) have decentralized the responsibility for the transparency of AI usage. Specifically, LLM users are now encouraged or required to disclose the use of LLM-generated content for varied types of…

Explainable models in Artificial Intelligence are often employed to ensure transparency and accountability of AI systems. The fidelity of the explanations are dependent upon the algorithms used as well as on the fidelity of the data. Many…

Training against white-box deception detectors has been proposed as a way to make AI systems honest. However, such training risks models learning to obfuscate their deception to evade the detector. Prior work has studied obfuscation only in…

The concept of disinformation is to use fake messages to confuse people in order to protect the real information. This strategy can be adapted into data science to protect valuable private and sensitive data. Huge amount of private data are…

Clustering algorithms rely on complex optimisation processes that may be difficult to comprehend, especially for individuals who lack technical expertise. While many explainable artificial intelligence techniques exist for supervised…

The goal of homomorphic encryption is to encrypt data such that another party can operate on it without being explicitly exposed to the content of the original data. We introduce an idea for a privacy-preserving transformation on natural…

We introduce a new class of attacks on machine learning models. We show that an adversary who can poison a training dataset can cause models trained on this dataset to leak significant private details of training points belonging to other…

Data mining services require accurate input data for their results to be meaningful, but privacy concerns may influence users to provide spurious information. To encourage users to provide correct inputs, we recently proposed a data…

The stochastic nature of artificial intelligence (AI) models introduces risk to business applications that use AI models without careful consideration. This paper offers an approach to use AI techniques to gain insights on the usage of the…

Users often ask dialogue systems ambiguous questions that require clarification. We show that current language models rarely ask users to clarify ambiguous questions and instead provide incorrect answers. To address this, we introduce CLAM:…

Privacy and transparency are two key foundations of trustworthy machine learning. Model explanations offer insights into a model's decisions on input data, whereas privacy is primarily concerned with protecting information about the…

In this paper, we describe our approach to achieve distributed differential privacy by sampling alone. Our mechanism works in the semi-honest setting (honest-but-curious whereby aggregators attempt to peek at the data though follow the…

Cautious predictions -- where a machine learning model abstains when uncertain -- are crucial for limiting harmful errors in safety-critical applications. In this work, we identify a novel threat: a dishonest institution can exploit these…

To handle underspecified or ambiguous queries, AI assistants need a policy for managing their uncertainty to determine (a) when to guess the user intent and answer directly, (b) when to enumerate and answer multiple possible intents, and…

Detecting semantic backdoors in classification models--where some classes can be activated by certain natural, but out-of-distribution inputs--is an important problem that has received relatively little attention. Semantic backdoors are…

As artificial intelligence (AI) assistants become more widely adopted in safety-critical domains, it becomes important to develop safeguards against potential failures or adversarial attacks. A key prerequisite to developing these…