Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks
☆32Jul 9, 2024Updated last year
Alternatives and similar repositories for SafeUnlearning
Users that are interested in SafeUnlearning are comparing it to the libraries listed below
Sorting:
- [ACL 2025] LongSafety: Evaluating Long-Context Safety of Large Language Models☆16Jun 18, 2025Updated 8 months ago
- [ICLR 2026] BARREL: Boundary-Aware Reasoning for Factual and Reliable LRMs☆17May 21, 2025Updated 9 months ago
- [ACL 2024] Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization☆29Jul 9, 2024Updated last year
- [NeurIPS 2024] "Self-Calibrated Tuning of Vision-Language Models for Out-of-Distribution Detection"☆13Oct 28, 2024Updated last year
- Focused Papers, Delivered Simply ：）☆50Dec 25, 2025Updated 2 months ago
- [NeurIPS'22] Trap and Replace: Defending Backdoor Attacks by Trapping Them into an Easy-to-Replace Subnetwork. Haotao Wang, Junyuan Hong,…☆15Nov 27, 2023Updated 2 years ago
- This is the oficial repository for "Safer-Instruct: Aligning Language Models with Automated Preference Data"☆17Feb 22, 2024Updated 2 years ago
- Official repository for ICML 2024 paper "On Prompt-Driven Safeguarding for Large Language Models"☆107May 20, 2025Updated 9 months ago
- ☆21Sep 16, 2024Updated last year
- ☆23Mar 21, 2025Updated 11 months ago
- Implementation for <Robust Weight Perturbation for Adversarial Training> in IJCAI'22.☆16Jul 1, 2022Updated 3 years ago
- ☆18Dec 10, 2022Updated 3 years ago
- ☆25Mar 16, 2025Updated 11 months ago
- Camouflage poisoning via machine unlearning☆19Jul 3, 2025Updated 7 months ago
- Code for paper: PoisonPrompt: Backdoor Attack on Prompt-based Large Language Models, IEEE ICASSP 2024. Demo//124.220.228.133:11107☆20Aug 10, 2024Updated last year
- [ACL 2025] Data and Code for Paper VLSBench: Unveiling Visual Leakage in Multimodal Safety☆54Jul 21, 2025Updated 7 months ago
- ☆19Jun 21, 2021Updated 4 years ago
- [NeurIPS 2021] Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training☆32Jan 9, 2022Updated 4 years ago
- This dataset contains results from all rounds of Adversarial Nibbler. This data includes adversarial prompts fed into public generative t…☆25Feb 3, 2025Updated last year
- Converting Mixtral-8x7B to Mixtral-[1~7]x7B☆22Mar 4, 2024Updated last year
- Not All Poisons are Created Equal: Robust Training against Data Poisoning (ICML 2022)☆22Aug 8, 2022Updated 3 years ago
- [NeurIPS 2022] "Randomized Channel Shuffling: Minimal-Overhead Backdoor Attack Detection without Clean Datasets" by Ruisi Cai*, Zhenyu Zh…☆21Oct 1, 2022Updated 3 years ago
- ☆21Oct 25, 2023Updated 2 years ago
- Code for Neurips 2024 paper "Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models"☆59Jan 15, 2025Updated last year
- ☆25Jun 16, 2024Updated last year
- Official Repository for ACL 2024 Paper SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding☆151Jul 19, 2024Updated last year
- ☆25Feb 27, 2023Updated 3 years ago
- This is the official code for the paper "Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning" (NeurIPS2024)☆25Sep 10, 2024Updated last year
- ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors [EMNLP 2024 Findings]☆226Sep 29, 2024Updated last year
- ☆29Jul 1, 2023Updated 2 years ago
- ☆26Jun 5, 2024Updated last year
- Code repo for the model organisms and convergent directions of EM papers.☆51Sep 22, 2025Updated 5 months ago
- ☆29Jan 16, 2023Updated 3 years ago
- The official repository of the paper "On the Exploitability of Instruction Tuning".☆70Feb 5, 2024Updated 2 years ago
- Analyze AI agent trajectories: extract actions, summarize, embed, and visualize.☆93Feb 20, 2026Updated last week
- ☆35Feb 20, 2025Updated last year
- ☆125Nov 13, 2023Updated 2 years ago
- [ACL 2024] Code and data for "Machine Unlearning of Pre-trained Large Language Models"☆66Sep 30, 2024Updated last year
- Implementation of "Adversarial purification with Score-based generative models", ICML 2021☆30Oct 24, 2021Updated 4 years ago