【清华出品】NLP新方向文本对抗攻击与防御必读论文列表

【导读】深度网络在精心设计的对抗攻击样例面前是非常脆弱的，这使得研究对抗攻击与防御的工作，特别是研究CV上的对抗攻击和防御的工作如雨后春笋，节节攀升。显然，对抗攻击这个事，在使用深度网络的所有方向上都会有，例如：图像、文本、语音等等。NLP上的对抗攻击和防御，也顺势而起，虽然目前的相关工作，只有近期零星的几篇，但是已经引起了NLP社区的浓厚的研究兴趣。前段时间，清华NLP组的同学们，在GitHub上开源了一个list，名为：文本对抗攻击与防御必读paper列表，相信对文本对抗攻击和防御感兴趣的同学，看完列表的文章，一定会有所收获。

GitHub库：

TAADpapers

GitHub地址：

https://github.com/thunlp/TAADpapers

作者：

Chenghao Yang, Fanchao Qi and Yuan Zang

题目：

必读的文本对抗攻击与防御文章

Must-read Papers on Textual Adversarial Attack and Defense (TAAD)

Section	Description
Survey	Survey papers on Textual Attack and Defense
Black-box Only	Attack generators only have access to confidence of victim models
White-box Only	Attack generators have full access to victim models
Both	Papers work on both black-box and white-box setting
Defense Only	Papers work on defense
Evaluation	Papers propose new evaluations of textual attacks and defense
Application of TAAD in Other Fields	Papers apply TAAD in other fields except Natural Language Processing (NLP)

Survey Papers：

Analysis Methods in Neural Language Processing: A Survey. Yonatan Belinkov, James Glass. TACL 2019. [pdf]
Towards a Robust Deep Neural Network in Text Domain A Survey. Wenqi Wang, Lina Wang, Benxiao Tang, Run Wang, Aoshuang Ye. 2019. [pdf]
Adversarial Attacks on Deep Learning Models in Natural Language Processing: A Survey. Wei Emma Zhang, Quan Z. Sheng, Ahoud Alhazmi, Chenliang Li. 2019. [pdf]

Black-box Only:

PAWS: Paraphrase Adversaries from Word Scrambling. Yuan Zhang, Jason Baldridge, Luheng He. NAACL-HLT 2019. [pdf] [dataset]
Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systems. Steffen Eger, Gözde Gül ¸Sahin, Andreas Rücklé, Ji-Ung Lee, Claudia Schulz, Mohsen Mesgar, Krishnkant Swarnkar, Edwin Simpson, Iryna Gurevych.NAACL-HLT 2019. [pdf] [code&data]
Adversarial Over-Sensitivity and Over-Stability Strategies for Dialogue Models. Tong Niu, Mohit Bansal. CoNLL 2018. [pdf] [code&data]
Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledge. Pasquale Minervini, Sebastian Riedel. CoNLL 2018. [pdf] [code&data]
Generating Natural Language Adversarial Examples. Moustafa Alzantot, Yash Sharma, Ahmed Elgohary, Bo-Jhang Ho, Mani Srivastava, Kai-Wei Chang. EMNLP 2018. [pdf] [code]
Breaking NLI Systems with Sentences that Require Simple Lexical Inferences. Max Glockner, Vered Shwartz, Yoav Goldberg ACL 2018. [pdf] [dataset]
AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples. Dongyeop Kang, Tushar Khot, Ashish Sabharwal, Eduard Hovy. ACL 2018. [pdf] [code]
Semantically Equivalent Adversarial Rules for Debugging NLP Models. Marco Tulio Ribeiro, Sameer Singh, Carlos Guestrin ACL 2018. [pdf] [code]
Robust Machine Comprehension Models via Adversarial Training. Yicheng Wang, Mohit Bansal. NAACL-HLT 2018. [pdf] [dataset]
Adversarial Example Generation with Syntactically Controlled Paraphrase Networks. Mohit Iyyer, John Wieting, Kevin Gimpel, Luke Zettlemoyer. NAACL-HLT 2018. [pdf] [code&data]
Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers. Ji Gao, Jack Lanchantin, Mary Lou Soffa, Yanjun Qi. IEEE SPW 2018. [pdf] [code]
Synthetic and Natural Noise Both Break Neural Machine Translation. Yonatan Belinkov, Yonatan Bisk. ICLR 2018. [pdf] [code&data]
Generating Natural Adversarial Examples. Zhengli Zhao, Dheeru Dua, Sameer Singh. ICLR 2018. [pdf] [code]
Adversarial Examples for Evaluating Reading Comprehension Systems. Robin Jia, and Percy Liang. EMNLP 2017. [pdf] [code]
Adversarial Sets for Regularising Neural Link Predictors. Pasquale Minervini, Thomas Demeester, Tim Rocktäschel, Sebastian Riedel. UAI 2017 [pdf] [code]

White-box Only:

On Adversarial Examples for Character-Level Neural Machine Translation. Javid Ebrahimi, Daniel Lowd, Dejing Dou. COLING 2018. [pdf] [code]
HotFlip: White-Box Adversarial Examples for Text Classification. Javid Ebrahimi, Anyi Rao, Daniel Lowd, Dejing Dou. ACL 2018. [pdf] [code]
Towards Crafting Text Adversarial Samples. Suranjana Samanta, Sameep Mehta. ECIR 2018. [pdf]

Both:

TEXTBUGGER: Generating Adversarial Text Against Real-world Applications. Jinfeng Li, Shouling Ji, Tianyu Du, Bo Li, Ting Wang. NDSS 2019. [pdf]
Comparing Attention-based Convolutional and Recurrent Neural Networks: Success and Limitations in Machine Reading Comprehension. Matthias Blohm, Glorianna Jagfeld, Ekta Sood, Xiang Yu, Ngoc Thang Vu. CoNLL 2018. [pdf]
Deep Text Classification Can be Fooled. Bin Liang, Hongcheng Li, Miaoqiang Su, Pan Bian, Xirong Li, Wenchang Shi.IJCAI 2018. [pdf]

Defense Only:

Combating Adversarial Misspellings with Robust Word Recognition. Danish Pruthi, Bhuwan Dhingra, Zachary C. Lipton. ACL 2019. [pdf] [code]

Evaluation:

On Evaluation of Adversarial Perturbations for Sequence-to-Sequence Models. Paul Michel, Xian Li, Graham Neubig, Juan Miguel Pino. NAACL-HLT 2019. [pdf] [code]

Application of TAAD in Other Fields:

Unified Visual-Semantic Embeddings: Bridging Vision and Language with Structured Meaning Representations.Hao Wu, Jiayuan Mao, Yufeng Zhang, Yuning Jiang, Lei Li, Weiwei Sun, Wei-Ying Ma. CVPR 2019. [pdf]
Learning Visually-Grounded Semantics from Contrastive Adversarial Samples. Haoyue Shi, Jiayuan Mao, Tete Xiao, Yuning Jiang, Jian Sun. COLING 2018. [pdf] [code]

-END-

专 · 知

专知，专业可信的人工智能知识分发，让认知协作更快更好！欢迎登录www.zhuanzhi.ai，注册登录专知，获取更多AI知识资料！