Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution

Qi, Fanchao; Yao, Yuan; Xu, Sophia; Liu, Zhiyuan; Sun, Maosong

doi:10.18653/v1/2021.acl-long.377

Cited by 65 publications

(40 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sememe knowledge bases like HowNet (Dong and Dong, 2006) use a set of predefined sememes to annotate words, so that the meaning of a word can be precisely expressed by its sememes. With the help of such sememe knowledge bases, sememes have been successfully utilized in various NLP tasks (Qi et al, 2021a), including semantic composition (Qi et al, 2019), word sense disambiguation (Hou et al, 2020), reverse dictionary (Zhang et al, 2020a), backdoor learning (Qi et al, 2021b), etc.…”

Section: Incorporation Of Sememesmentioning

confidence: 99%

QuoteR: A Benchmark of Quote Recommendation for Writing

Qi¹,

Yang²,

Yi³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

It is very common to use quotations (quotes) to make our writings more elegant or convincing. To help people find appropriate quotes more efficiently, the task of quote recommendation is presented, aiming to recommend quotes that fit the current context of writing. There have been various quote recommendation approaches, but they are evaluated on different unpublished datasets. To facilitate the research on this task, we build a large and fully open quote recommendation dataset called QuoteR, which comprises three parts including English, standard Chinese and classical Chinese. Any part of it is larger than previous unpublished counterparts. We conduct an extensive evaluation of existing quote recommendation methods on QuoteR. Furthermore, we propose a new quote recommendation model that significantly outperforms previous methods on all three parts of QuoteR. All the code and data of this paper are available at https: //github.com/thunlp/QuoteR.

show abstract

Section: Incorporation Of Sememesmentioning

confidence: 99%

QuoteR: A Benchmark of Quote Recommendation for Writing

Qi¹,

Yang²,

Yi³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…They are usually very natural and fluent, thus barely distinguishable from normal samples. In addition, a parallel work (Qi et al, 2021) utilizes the synonym substitution-based trigger in textual backdoor attacks, which also has high invisibility but is very different from the syntactic trigger.…”

Section: Backdoor Attacksmentioning

confidence: 99%

Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger

Li²,

Chen³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Backdoor attacks are a kind of insidious security threat against machine learning models. After being injected with a backdoor in training, the victim model will produce adversaryspecified outputs on the inputs embedded with predesigned triggers but behave properly on normal inputs during inference. As a sort of emergent attack, backdoor attacks in natural language processing (NLP) are investigated insufficiently. As far as we know, almost all existing textual backdoor attack methods insert additional contents into normal samples as triggers, which causes the trigger-embedded samples to be detected and the backdoor attacks to be blocked without much effort. In this paper, we propose to use the syntactic structure as the trigger in textual backdoor attacks. We conduct extensive experiments to demonstrate that the syntactic trigger-based attack method can achieve comparable attack performance (almost 100% success rate) to the insertionbased methods but possesses much higher invisibility and stronger resistance to defenses. These results also reveal the significant insidiousness and harmfulness of textual backdoor attacks. All the code and data of this paper can be obtained at https://github.com/ thunlp/HiddenKiller.

show abstract

“…'mm', 'bb' and 'James Bond', that can then be easily detected at test time. In [114] and [115], a less detectable trigger is used by relying on a proper combination of synonyms and syntaxes.…”

Section: B Extension To Domains Other Than Computer Visionmentioning

confidence: 99%

An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defences

Guo¹,

Tondi²,

Barni³

2021

Preprint

View full text Add to dashboard Cite

Together with impressive advances touching every aspect of our society, AI technology based on Deep Neural Networks (DNN) is bringing increasing security concerns. While attacks operating at test time have monopolised the initial attention of researchers, backdoor attacks, exploiting the possibility of corrupting DNN models by interfering with the training process, represents a further serious threat undermining the dependability of AI techniques. In a backdoor attack, the attacker corrupts the training data so to induce an erroneous behaviour at test time. Test time errors, however, are activated only in the presence of a triggering event corresponding to a properly crafted input sample. In this way, the corrupted network continues to work as expected for regular inputs, and the malicious behaviour occurs only when the attacker decides to activate the backdoor hidden within the network. In the last few years, backdoor attacks have been the subject of an intense research activity focusing on both the development of new classes of attacks, and the proposal of possible countermeasures. The goal of this overview paper is to review the works published until now, classifying the different types of attacks and defences proposed so far. The classification guiding the analysis is based on the amount of control that the attacker has on the training process, and the capability of the defender to verify the integrity of the data used for training, and to monitor the operations of the DNN at training and test time. As such, the proposed analysis is particularly suited to highlight the strengths and weaknesses of both attacks and defences with reference to the application scenarios they are operating in.

show abstract

Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution

Cited by 65 publications

References 45 publications

QuoteR: A Benchmark of Quote Recommendation for Writing

QuoteR: A Benchmark of Quote Recommendation for Writing

Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger

An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defences

Contact Info

Product

Resources

About