thunlp / ONION
Official implementation of the EMNLP 2021 paper "ONION: A Simple and Effective Defense Against Textual Backdoor Attacks"
☆29Updated 3 years ago
Related projects ⓘ
Alternatives and complementary repositories for ONION
- Code and data of the ACL-IJCNLP 2021 paper "Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger"☆37Updated 2 years ago
- Code and data of the EMNLP 2021 paper "Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer"☆41Updated 2 years ago
- Code for the paper "RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Models" (EMNLP 2021)☆22Updated 3 years ago
- Implementation of the paper "Exploring the Universal Vulnerability of Prompt-based Learning Paradigm" on Findings of NAACL 2022☆27Updated 2 years ago
- ☆23Updated 3 years ago
- Code for the paper "Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Models" (NAACL-…☆37Updated 3 years ago
- Code for the paper "Rethinking Stealthiness of Backdoor Attack against NLP Models" (ACL-IJCNLP 2021)☆21Updated 2 years ago
- ☆14Updated 3 years ago
- Natural Universal Trigger Search (NUTS)☆21Updated 3 years ago
- ☆17Updated 2 years ago
- Natural Language Attacks in a Hard Label Black Box Setting.☆43Updated 3 years ago
- An open-source toolkit for textual backdoor attack and defense (NeurIPS 2022 D&B, Spotlight)☆151Updated last year
- codes for "Searching for an Effective Defender:Benchmarking Defense against Adversarial Word Substitution"☆31Updated last year
- Code for the paper "BadPrompt: Backdoor Attacks on Continuous Prompts"☆36Updated 4 months ago
- Code and data of the ACL 2021 paper "Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution"☆16Updated 3 years ago
- Contextualized Perturbation for Textual Adversarial Attack, NAACL 2021☆43Updated 3 years ago
- Code for Findings-EMNLP 2023 paper: Multi-step Jailbreaking Privacy Attacks on ChatGPT☆25Updated last year
- A2T: Towards Improving Adversarial Training of NLP Models (EMNLP 2021 Findings)☆24Updated 3 years ago
- Generating Natural Language Adversarial Examples through Probability Weighted Word Saliency☆67Updated last year
- ☆9Updated 3 years ago
- ACL 2021 - Defense against Adversarial Attacks in NLP via Dirichlet Neighborhood Ensemble☆15Updated last year
- For Certified Robustness to Text Adversarial Attacks by Randomized [MASK]☆15Updated last month
- SAFER: A Structure-free Approach For cErtified Robustness to Adversarial Word Substitutions (ACL 2020)☆27Updated 3 years ago
- Official Code for ACL 2023 paper: "Ethicist: Targeted Training Data Extraction Through Loss Smoothed Soft Prompting and Calibrated Confid…☆23Updated last year
- Mostly recording papers about models' trustworthy applications. Intending to include topics like model evaluation & analysis, security, c…☆20Updated last year
- Hidden backdoor attack on NLP systems☆46Updated 3 years ago
- [ACL2024-Main] Data and Code for WaterBench: Towards Holistic Evaluation of LLM Watermarks☆18Updated last year
- Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models☆23Updated last year
- Unofficial implementation of "Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection"☆14Updated 4 months ago
- ☆23Updated last year