qingjiesjtu / USC
This is the code repository of our submission: Understanding the Dark Side of LLMs’ Intrinsic Self-Correction.
☆56Updated 4 months ago
Alternatives and similar repositories for USC:
Users that are interested in USC are comparing it to the libraries listed below
- Comprehensive Assessment of Trustworthiness in Multimodal Foundation Models☆18Updated last month
- Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attacks (IEEE S&P 2024)☆33Updated 2 weeks ago
- 🔥🔥🔥Breaking long thought processes of o1-like LLMs, such as DeepSeek-R1, QwQ☆28Updated last month
- This repository is the official implementation of the paper "ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning…☆17Updated last year
- A toolbox for backdoor attacks.☆21Updated 2 years ago
- ☆51Updated 4 months ago
- Code for paper: PoisonPrompt: Backdoor Attack on Prompt-based Large Language Models, IEEE ICASSP 2024. Demo//124.220.228.133:11107☆17Updated 8 months ago
- Code for NeurIPS 2024 Paper "Fight Back Against Jailbreaking via Prompt Adversarial Tuning"☆12Updated 6 months ago
- [ACL 2024] CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion☆38Updated 6 months ago
- [ACL 2024] Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization☆22Updated 9 months ago
- ☆26Updated 6 months ago
- Codes for NeurIPS 2021 paper "Adversarial Neuron Pruning Purifies Backdoored Deep Models"☆57Updated last year
- The official implementation of USENIX Security'23 paper "Meta-Sift" -- Ten minutes or less to find a 1000-size or larger clean subset on …☆18Updated 2 years ago
- Official Code for "Baseline Defenses for Adversarial Attacks Against Aligned Language Models"☆23Updated last year
- Repository for Towards Codable Watermarking for Large Language Models☆36Updated last year
- This is the code repository for "Uncovering Safety Risks of Large Language Models through Concept Activation Vector"☆36Updated 5 months ago
- Official Implementation of ICLR 2022 paper, ``Adversarial Unlearning of Backdoors via Implicit Hypergradient''☆53Updated 2 years ago
- ☆41Updated last month
- ☆24Updated 2 years ago
- Official Code for ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users (NeurIPS 2024)☆15Updated 6 months ago
- Simulator.☆104Updated 2 weeks ago
- ☆15Updated 2 years ago
- ☆81Updated 3 years ago
- Backdoor Safety Tuning (NeurIPS 2023 & 2024 Spotlight)☆25Updated 5 months ago
- Code repo of our paper Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis (https://arxiv.org/abs/2406.10794…☆19Updated 9 months ago
- ☆18Updated 7 months ago
- MASTERKEY is a framework designed to explore and exploit vulnerabilities in large language model chatbots by automating jailbreak attacks…☆21Updated 7 months ago
- A curated list of trustworthy Generative AI papers. Daily updating...☆71Updated 8 months ago
- [NeurIPS 2024] Fight Back Against Jailbreaking via Prompt Adversarial Tuning☆10Updated 6 months ago
- This is an official repository for Practical Membership Inference Attacks Against Large-Scale Multi-Modal Models: A Pilot Study (ICCV2023…☆22Updated last year