Data Poisoning for In-context Learning
Authors:
Open Access
- 1 January 2025
- conference paper
- Published by Association for Computational Linguistics (ACL)
- p. 1680-1700
Abstract
Pengfei He, Han Xu, Yue Xing, Hui Liu, Makoto Yamada, Jiliang Tang. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning2024
- Fraud Detection under Siege: Practical Poisoning Attacks and Defense StrategiesACM Transactions on Privacy and Security, 2023
- In-Context Learning Creates Task Vectors2023
- Active Learning Principles for In-Context Learning with Large Language Models2023
- Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models2023
- NOTABLE: Transferable Backdoor Attacks Against Prompt-based NLP Models2023
- A Survey on Multi-Task LearningIEEE Transactions on Knowledge and Data Engineering, 2021
- Poison Attacks against Text Datasets with Conditional Adversarially Regularized Autoencoder2020
- BadNets: Evaluating Backdooring Attacks on Deep Neural NetworksIEEE Access, 2019
- TextBugger: Generating Adversarial Text Against Real-world Applications2019