相关论文: SoK: Content Moderation for End-to-End Encryption

Emails today are often encrypted, but only between mail servers---the vast majority of emails are exposed in plaintext to the mail servers that handle them. While better than no encryption, this arrangement leaves open the possibility of…

Online toxic content has grown into a pervasive phenomenon, intensifying during times of crisis, elections, and social unrest. A significant amount of research has been focused on detecting or analyzing toxic content using machine-learning…

The inclusion of pervasive computing devices in a democratized edge computing ecosystem can significantly expand the capability and coverage of near-end computing for large-scale applications. However, offloading user tasks to heterogeneous…

The growth of online platforms and user content requires strong content moderation systems that can handle complex inputs from various media types. While large language models (LLMs) are effective, their high computational cost and latency…

Social media platforms increasingly employ proactive moderation techniques, such as detecting and curbing toxic and uncivil comments, to prevent the spread of harmful content. Despite these efforts, such approaches are often criticized for…

The European Union introduced the Digital Services Act (DSA) to address the risks associated with digital platforms and promote a safer online environment. However, despite the potential of components such as the Transparency Database,…

In today's digital age, the imperative to protect data privacy and security is a paramount concern, especially for business-to-business (B2B) enterprises that handle sensitive information. These enterprises are increasingly constructing…

End-to-end (E2E) models, which directly predict output character sequences given input speech, are good candidates for on-device speech recognition. E2E models, however, present numerous challenges: In order to be truly useful, such models…

With the rising popularity of the internet and the widespread use of networks and information systems via the cloud and data centers, the privacy and security of individuals and organizations have become extremely crucial. In this…

Watermarking embeds information into digital content like images, audio, or text, imperceptible to humans but robustly detectable by specific algorithms. This technology has important applications in many challenges of the industry such as…

Artificial intelligence (AI) increasingly powers sensitive applications in domains such as healthcare and finance, relying on both linear operations (e.g., matrix multiplications in large language models) and non-linear operations (e.g.,…

Internet of Things (IoT) is an innovative paradigm envisioned to provide massive applications that are now part of our daily lives. Millions of smart devices are deployed within complex networks to provide vibrant functionalities including…

Text embeddings are fundamental to many natural language processing (NLP) tasks, extensively applied in domains such as recommendation systems and information retrieval (IR). Traditionally, transmitting embeddings instead of raw text has…

In software development, privacy preservation has become essential with the rise of privacy concerns and regulations such as GDPR and CCPA. While several tools, guidelines, methods, methodologies, and frameworks have been proposed to…

Large language models (LLMs) are sophisticated artificial intelligence systems that enable machines to generate human-like text with remarkable precision. While LLMs offer significant technological progress, their development using vast…

Machine learning (ML) is widely used to moderate online content. Despite its scalability relative to human moderation, the use of ML introduces unique challenges to content moderation. One such challenge is predictive multiplicity: multiple…

This paper considers the problem of simultaneously communicating two messages, a high-security message and a low-security message, to a legitimate receiver, referred to as the security embedding problem. An information-theoretic formulation…

Data sent over the Internet can be monitored and manipulated by intermediate entities in the data path from the source to the destination. For unencrypted communications (and some encrypted communications with known weaknesses),…

Speech recognition applications cover a range of different audio and text distributions, with different speaking styles, background noise, transcription punctuation and character casing. However, many speech recognition systems require…

We introduce a model of content moderation for sale, where a platform can channel attention in two ways: direct steering that makes content visible to consumers and certification that controls what consumers know about the content. The…