yihuaihong / ConceptVectors
ConceptVectors Benchmark and Code for the paper "Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces"
☆14Updated 2 weeks ago
Related projects: ⓘ
- ☆37Updated 10 months ago
- Is In-Context Learning Sufficient for Instruction Following in LLMs?☆19Updated 3 months ago
- ☆47Updated last year
- Parsimonious Concept Engineering (PaCE) uses sparse coding on a large-scale concept dictionary to effectively improve the trustworthiness…☆25Updated 3 months ago
- [Arxiv 2024] Adversarial attacks on multimodal agents☆33Updated 2 months ago
- ☆30Updated last year
- Codebase for decoding compressed trust.☆20Updated 4 months ago
- Model Editing Can Hurt General Abilities of Large Language Models☆29Updated 7 months ago
- [ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning☆79Updated 3 months ago
- ☆21Updated 2 months ago
- Official Repository for Dataset Inference for LLMs☆21Updated last month
- ☆23Updated 4 months ago
- Code of paper "Accelerating Greedy Coordinate Gradient via Probe Sampling"☆16Updated 5 months ago
- [ACL 2024] Code and data for "Machine Unlearning of Pre-trained Large Language Models"☆34Updated 4 months ago
- Official Repository for The Paper: Safety Alignment Should Be Made More Than Just a Few Tokens Deep☆22Updated 2 months ago
- [ICML 2023] "Robust Weight Signatures: Gaining Robustness as Easy as Patching Weights?" by Ruisi Cai, Zhenyu Zhang, Zhangyang Wang☆16Updated last year
- Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks☆19Updated 2 months ago
- Code for safety test in "Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates"☆16Updated 6 months ago
- ☆15Updated 2 months ago
- The official repository of the paper "On the Exploitability of Instruction Tuning".☆56Updated 7 months ago
- Official code implementation of SKU, Accepted by ACL 2024 Findings☆11Updated 4 months ago
- [ICLR 2024] Provable Robust Watermarking for AI-Generated Text☆25Updated 9 months ago
- Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NextGenAISafety @ ICML 2024)☆37Updated last month
- [ACL 2024] CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion☆14Updated this week
- Weak-to-Strong Jailbreaking on Large Language Models☆62Updated 6 months ago
- AmpleGCG: Learning a Universal and Transferable Generator of Adversarial Attacks on Both Open and Closed LLM☆33Updated 3 weeks ago
- ☆57Updated last week
- Codebase for Inference-Time Policy Adapters☆19Updated 10 months ago
- This is the official code for the paper "Vaccine: Perturbation-aware Alignment for Large Language Models"☆12Updated last week
- Official implementation of Privacy Implications of Retrieval-Based Language Models (EMNLP 2023). https://arxiv.org/abs/2305.14888☆36Updated 3 months ago