Adversarial Attacks on Deep Learning Models in Natural Language Processing: A Survey

By W. Zhang, Quan.Z Sheng, Ahoud Abdulrahmn F. Alhazmi, Chenliang Li. 2019

601 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Mohit Bansal
2. Eric Wallace
3. Yonatan Belinkov
4. Percy Liang
5. Dacheng Tao
6. Zhe Gan
7. Jianfeng Gao
8. Douwe Kiela

Citations in the corpus

• Make Up Your Mind! Adversarial Generation of Inconsistent Natural Language Explanations (2019, background): “Generating adversarial examples has received increasing attention in natural language processing [31, 30].”
• Mitigating Gender Bias for Neural Dialogue Generation with Adversarial Learning (2020, background): “Dialogue systems have been shown to be sensitive to the input messages (Niu and Bansal, 2018; Zhang et al., 2020; Xu et al., 2020).”
• DynaSent: A Dynamic Benchmark for Sentiment Analysis (2020, background): “It has, however, entered into a more challenging phase for research and technology development: while present-day models achieve outstanding results on all available benchmark tasks, they still fall short when deployed as part of real-world systems (Burn-Murdoch, 2013; Grimes, 2014, 2017; Gossett, 2”
• Context Dependent Semantic Parsing: A Survey (2020, background): “Another potential benefit of understanding causal structures in context is to improve robustness of parsers by ignoring non-robust features (Zhang et al., 2020; Ilyas et al., 2019).”
• Large Language Model Alignment: A Survey (2023, methodology): “Adversarial attacks are techniques employed to compromise the performance or behavior of machine learning models, particularly deep learning models, by introducing small and carefully crafted perturbations to the input data (Akhtar and Mian, 2018; Zhang et al., 2020; Qiu et al., 2022; Goyal et al., ”
• LIFT: Language-Interfaced Fine-Tuning for Non-Language Machine Learning Tasks (2022, methodology): “We apply transfer attack [51] since we do not have full access to the GPT-3 model, and finding adversarial examples in the discrete input space is complex [52].”
• On Robust Prefix-Tuning for Text Classification (2022, background): “In the NLP community, a variety of techniques for generating adversarial examples have been proposed to attack a text classifier by perturbing inputs (Zhang et al., 2020).”
• ReCode: Robustness Evaluation of Code Generation Models (2022, background): “Other existing works on robustness in text or code tasks have focused on classification and are not directly applicable to code generation (Zhang et al., 2020; Jha and Reddy, 2022).”
• Towards Understanding Neural Machine Translation with Word Importance (2019, unknown): “Since the perturbation on text is notoriously hard (Zhang et al., 2019) due to the semantic shifting problem, in this exper-iment, we investigate three types of perturbations to avoid the potential bias : • Deletion perturbation removes the selected words from the input sentence, and it can be…”
• Evaluating the Robustness of Neural Language Models to Input Perturbations (2021, background): “Some types of perturbation utilized in this work were already tested in previous work on adversarial attacks on NLP systems (Zeng et al., 2020; Zhang et al., 2020).”
• SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks (2023, unknown): “Over the last few years, an amalgamation of attacks and defenses have been proposed in the literature surrounding the robustness of language models [60, 61].”
• Contextualized Perturbation for Textual Adversarial Attack (2020, background): “…lexical substitution (Wu et al., 2019a; Zhou et al., 2019a; Qiang et al., 2020; Wu et al., 2019b, inter alia ) to non-autoregressive generation (Gu et al., 2018; Lee et al., 2018; Ghazvininejad et al., 2019; Ma et al., 2019; Sun et al., 2019; Ren et al., 2020; Zhang et al., 2020b, inter alia ).”
• Jailbroken: How Does LLM Safety Training Fail? (2023, background): “There also exists an extensive literature on adversarial examples for deep learning models in natural language processing and elsewhere (see Chakraborty et al. [15] and Zhang et al. [57] for surveys).”
• Universal Adversarial Triggers for Attacking and Analyzing NLP (2019, background): “We turn the reader to Zhang et al. (2019) for a recent survey.”
• Adversarial Language Games for Advanced Natural Language Intelligence (2019, background): “(Zhang et al., 2019) aims at finding adversarial examples of NLP models to evaluate the model robustness (Jia and Liang, 2017; Michel et al., 2019) or to help to train a robust model (Wang and Bansal, 2018; Cheng et al., 2019b).”
• Evaluation Gaps in Machine Learning Practice (2022, methodology): “Borrowing adversarial techniques from security testing and privacy testing, adversarial testing of models requires due diligence to trigger the most harmful model predictions, using either manually chosen or algorithmically generated test instances [52, 144, 176, 178].”
• A Survey on Vision Transformer (2020, background): “Although the robustness has been investigated in [277], [155], [157], it is still an open problem waiting to be solved.”
• It’s Morphin’ Time! Combating Linguistic Discrimination with Inflectional Perturbations (2020, background): “Zhang et al. (2019a) provide a comprehensive survey of this topic.”