Zeyang Sha
YOU?
Author Swipe
View article: Agent Safety Alignment via Reinforcement Learning
Agent Safety Alignment via Reinforcement Learning Open
The emergence of autonomous Large Language Model (LLM) agents capable of tool usage has introduced new safety risks that go beyond traditional conversational misuse. These agents, empowered to execute external functions, are vulnerable to …
View article: A Survey of LLM-Driven AI Agent Communication: Protocols, Security Risks, and Defense Countermeasures
A Survey of LLM-Driven AI Agent Communication: Protocols, Security Risks, and Defense Countermeasures Open
In recent years, Large-Language-Model-driven AI agents have exhibited unprecedented intelligence and adaptability. Nowadays, agents are undergoing a new round of evolution. They no longer act as an isolated island like LLMs. Instead, they …
View article: Can VLMs Detect and Localize Fine-Grained AI-Edited Images?
Can VLMs Detect and Localize Fine-Grained AI-Edited Images? Open
Fine-grained detection and localization of localized image edits is crucial for assessing content authenticity, especially as modern diffusion models and image editors can produce highly realistic manipulations. However, this problem faces…
View article: ZeroFake: Zero-Shot Detection of Fake Images Generated and Edited by Text-to-Image Generation Models
ZeroFake: Zero-Shot Detection of Fake Images Generated and Edited by Text-to-Image Generation Models Open
The text-to-image generation model has attracted significant interest from both academic and industrial communities. These models can generate the images based on the given prompt descriptions. Their potent capabilities, while beneficial, …
View article: Games and Beyond: Analyzing the Bullet Chats of Esports Livestreaming
Games and Beyond: Analyzing the Bullet Chats of Esports Livestreaming Open
Esports, short for electronic sports, is a form of competition using video games and has attracted more than 530 million audiences worldwide. To watch esports, people utilize online livestreaming platforms. Recently, a novel interaction me…
View article: Prompt Stealing Attacks Against Large Language Models
Prompt Stealing Attacks Against Large Language Models Open
The increasing reliance on large language models (LLMs) such as ChatGPT in various fields emphasizes the importance of ``prompt engineering,'' a technology to improve the quality of model outputs. With companies investing significantly in …
View article: Reconstruct Your Previous Conversations! Comprehensively Investigating Privacy Leakage Risks in Conversations with GPT Models
Reconstruct Your Previous Conversations! Comprehensively Investigating Privacy Leakage Risks in Conversations with GPT Models Open
Significant advancements have recently been made in large language models represented by GPT models. Users frequently have multi-round private conversations with cloud-hosted GPT models for task optimization. Yet, this operational paradigm…
View article: Comprehensive Assessment of Toxicity in ChatGPT
Comprehensive Assessment of Toxicity in ChatGPT Open
Moderating offensive, hateful, and toxic language has always been an important but challenging topic in the domain of safe use in NLP. The emerging large language models (LLMs), such as ChatGPT, can potentially further accentuate this thre…
View article: Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders
Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders Open
Self-supervised representation learning techniques have been developing rapidly to make full use of unlabeled images. They encode images into rich features that are oblivious to downstream tasks. Behind their revolutionary representation p…
View article: From Visual Prompt Learning to Zero-Shot Transfer: Mapping Is All You Need
From Visual Prompt Learning to Zero-Shot Transfer: Mapping Is All You Need Open
Visual prompt learning, as a newly emerged technique, leverages the knowledge learned by a large-scale pre-trained model and adapts it to downstream tasks through the usage of prompts. While previous research has focused on designing effec…
View article: Fine-Tuning Is All You Need to Mitigate Backdoor Attacks
Fine-Tuning Is All You Need to Mitigate Backdoor Attacks Open
Backdoor attacks represent one of the major threats to machine learning models. Various efforts have been made to mitigate backdoors. However, existing defenses have become increasingly complex and often require high computational resource…
View article: DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models
DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models Open
Text-to-image generation models that generate images based on prompt descriptions have attracted an increasing amount of attention during the past few months. Despite their encouraging performance, these models raise concerns about the mis…
View article: Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders
Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders Open
Self-supervised representation learning techniques have been developing rapidly to make full use of unlabeled images. They encode images into rich features that are oblivious to downstream tasks. Behind their revolutionary representation p…