jiah-li / magicLinks
The repo for paper: Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models.
☆9Updated 6 months ago
Alternatives and similar repositories for magic
Users that are interested in magic are comparing it to the libraries listed below
Sorting:
- ☆14Updated last week
- ☆41Updated 8 months ago
- Our research proposes a novel MoGU framework that improves LLMs' safety while preserving their usability.☆15Updated 5 months ago
- ☆21Updated 5 months ago
- Official repository for the paper "Gradient-based Jailbreak Images for Multimodal Fusion Models" (https//arxiv.org/abs/2410.03489)☆17Updated 8 months ago
- Towards Safe LLM with our simple-yet-highly-effective Intention Analysis Prompting☆16Updated last year
- Code repo of our paper Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis (https://arxiv.org/abs/2406.10794…☆20Updated 11 months ago
- ☆21Updated 3 months ago
- Official code implementation of SKU, Accepted by ACL 2024 Findings☆15Updated 6 months ago
- ☆30Updated last year
- The official repository of 'Unnatural Language Are Not Bugs but Features for LLMs'☆19Updated last month
- Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NeurIPS 2024)☆61Updated 5 months ago
- ☆20Updated 6 months ago
- Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks☆28Updated 11 months ago
- Code for our paper "Defending ChatGPT against Jailbreak Attack via Self-Reminder" in NMI.☆50Updated last year
- [ACL 2024] Code and data for "Machine Unlearning of Pre-trained Large Language Models"☆59Updated 8 months ago
- Restore safety in fine-tuned language models through task arithmetic☆28Updated last year
- Code for paper "Universal Jailbreak Backdoors from Poisoned Human Feedback"☆55Updated last year
- Code for the paper "BadPrompt: Backdoor Attacks on Continuous Prompts"☆36Updated 11 months ago
- [ACL 2024] Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization☆24Updated 11 months ago
- Code for safety test in "Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates"☆18Updated last year
- EMNLP 2024: Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue☆35Updated last month
- [ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning☆94Updated last year
- Official repo for EMNLP'24 paper "SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning"☆26Updated 8 months ago
- Submission Guide + Discussion Board for AI Singapore Global Challenge for Safe and Secure LLMs (Track 1A).☆16Updated 11 months ago
- ☆44Updated 4 months ago
- Code for "Universal Adversarial Triggers Are Not Universal."☆17Updated last year
- Code for paper 'Batch-ICL: Effective, Efficient, and Order-Agnostic In-Context Learning'☆16Updated last year
- ConceptVectors Benchmark and Code for the paper "Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces"☆35Updated 4 months ago
- Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models☆29Updated last year