Paper Summary

Natural Language Processing

• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (NAACL 2019)

• Attention is All You Need (NIPS 2017)

• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding (ICLR 2019)

​

Adversarial Attack

• ONION: A Simple and Effective Defense Against Textual Backdoor Attacks (EMNLP 2021)

• BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Models (ICLR 2022)

• BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvements (ICML 2021 Workshop AML Program)

• Universal Adversarial Triggers for Attacking and Analyzing NLP (EMNLP 2019)

• Weight Poisoning Attacks on Pre-trained Models 

• Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger (ACL-IJCNLP 2021)

• The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networks

• Explaining and Harnessing Adversarial Examples (ICLR 2015) & Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment (AAAI 2020 Oral) [video] [slides]

  • Machine Learning course Final Presentation Project, Instructor: Prof. Dit-Yan YEUNG. ​

​