NISPLab / JBShield
Code for USENIX Security 2025 paper "JBShield: Defending Large Language Models from Jailbreak Attacks through Activated Concept Analysis and Manipulation"
☆177Updated last month
Alternatives and similar repositories for JBShield:
Users that are interested in JBShield are comparing it to the libraries listed below
- ☆167Updated 3 weeks ago
- ☆148Updated 8 months ago
- High-efficiency Secure Two Party Computation on GPU☆176Updated 3 weeks ago
- The implementation of our AAAI 2024 paper "Reverse Multi-Choice Dialogue Commonsense Inference with Graph-of-Thought"☆197Updated 3 weeks ago
- WHU大二 计算机设计 流水线CPU设计 课程作业☆13Updated last month
- ☆13Updated 3 weeks ago
- ☆25Updated last month
- ☆154Updated 3 weeks ago
- ☆148Updated 3 weeks ago
- SimdMSM: SIMD-accelerated Multi-Scalar Multiplication Framework for zkSNARKs☆167Updated this week
- MPC(Multi-Party Computation) all in one.☆148Updated 3 weeks ago
- ☆157Updated last month
- ☆147Updated 3 weeks ago
- ☆146Updated 3 weeks ago
- efficient anti side channel SHA3 algorithm software and hardware co-design☆155Updated this week
- ☆76Updated last week
- ☆149Updated last month
- ☆149Updated 3 weeks ago
- 基于IFTTT平台的隐私挖掘工具☆53Updated 3 weeks ago
- [开源软件发布]基于蓝牙的病毒追踪系统,采用BLE低功耗蓝牙,通过SM3加密认证保护用户数据安全性,提供包括Android开发,IOS开发,以及Java服务器开发的完整代码和直接可以运行的apk文件☆151Updated 3 weeks ago
- A secure IoT authentication framework based on hardware fingerprinting☆160Updated last month
- Code for CVPR 2024 "Revisiting Adversarial Training under Long-Tailed Distributions".☆158Updated last month
- 本项目基于兼具加密与计算双重能力的全同态加密算法、利用微软开源库Microsoft-Seal而设计出的一套能够保护医疗数据的云计算系统。☆59Updated 3 weeks ago
- 武汉大学国家网安院软件安全☆12Updated 4 months ago
- From Head to Tail: Efficient Black-box Model Inversion Attack via Long-tailed Learning - CVPR 2025☆12Updated last month
- This repository provide the studies on the security of language models for code (CodeLMs).☆49Updated 2 months ago
- 18级武汉大学国家网络安全学院暑期 实训备份☆11Updated 5 years ago
- WHU-武汉大学-操作系统概念-课程资料与习题解答☆27Updated 4 years ago
- ☆220Updated 11 months ago
- MASTERKEY is a framework designed to explore and exploit vulnerabilities in large language model chatbots by automating jailbreak attacks…☆20Updated 7 months ago