PyTorchKR Papers

최신 논문

총 98편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2306.05871

Towards a Robust Detection of Language Model Generated Text: Is ChatGPT that Easy to Detect?

Recent advances in natural language processing (NLP) have led to the development of large language models (LLMs) such as ChatGPT. This paper proposes a methodology for developing and evaluating ChatGPT detectors for French text, with a focus on investigating their robustness on out-of-domain data and against common attack schemes. The proposed method involves translating an English dataset into French and training a classifier on the translated data. Results show that the detectors can effectively detect ChatGPT-generated text, with a degree of robustness against basic attack techniques in in-domain settings. However, vulnerabilities are evident in out-of-domain contexts, highlighting the challenge of detecting adversarial text. The study emphasizes caution when applying in-domain testing results to a wider variety of content. We provide our translated datasets and models as open-source resources. https://gitlab.inria.fr/wantoun/robust-chatgpt-detection

arXiv:2306.01273공식 코드 · Papers with Code

VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformations

Adversarial attacks reveal serious flaws in deep learning models. More dangerously, these attacks preserve the original meaning and escape human recognition. Existing methods for detecting these attacks need to be trained using original/adversarial data. In this paper, we propose detection without training by voting on hard labels from predictions of transformations, namely, VoteTRANS. Specifically, VoteTRANS detects adversarial text by comparing the hard labels of input text and its transformation. The evaluation demonstrates that VoteTRANS effectively detects adversarial text across various state-of-the-art attacks, models, and datasets.

arXiv:2305.15587

How do humans perceive adversarial text? A reality check on the validity and naturalness of word-based adversarial attacks

Natural Language Processing (NLP) models based on Machine Learning (ML) are susceptible to adversarial attacks -- malicious algorithms that imperceptibly modify input text to force models into making incorrect predictions. However, evaluations of these attacks ignore the property of imperceptibility or study it under limited settings. This entails that adversarial perturbations would not pass any human quality gate and do not represent real threats to human-checked NLP systems. To bypass this limitation and enable proper assessment (and later, improvement) of NLP model robustness, we have surveyed 378 human participants about the perceptibility of text adversarial examples produced by state-of-the-art methods. Our results underline that existing text attacks are impractical in real-world scenarios where humans are involved. This contrasts with previous smaller-scale human studies, which reported overly optimistic conclusions regarding attack success. Through our work, we hope to position human perceptibility as a first-class success criterion for text attacks, and provide guidance for research to build effective attack algorithms and, in turn, design appropriate defence mechanisms.

arXiv:2305.13208

Iterative Adversarial Attack on Image-guided Story Ending Generation

Multimodal learning involves developing models that can integrate information from various sources like images and texts. In this field, multimodal text generation is a crucial aspect that involves processing data from multiple modalities and outputting text. The image-guided story ending generation (IgSEG) is a particularly significant task, targeting on an understanding of complex relationships between text and image data with a complete story text ending. Unfortunately, deep neural networks, which are the backbone of recent IgSEG models, are vulnerable to adversarial samples. Current adversarial attack methods mainly focus on single-modality data and do not analyze adversarial attacks for multimodal text generation tasks that use cross-modal information. To this end, we propose an iterative adversarial attack method (Iterative-attack) that fuses image and text modality attacks, allowing for an attack search for adversarial text and image in an more effective iterative way. Experimental results demonstrate that the proposed method outperforms existing single-modal and non-iterative multimodal attack methods, indicating the potential for improving the adversarial robustness of multimodal text generation models, such as multimodal machine translation, multimodal question answering, etc.

arXiv:2305.05095공식 코드 · Papers with Code

Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness

The CLIP (Contrastive Language-Image Pre-training) model and its variants are becoming the de facto backbone in many applications. However, training a CLIP model from hundreds of millions of image-text pairs can be prohibitively expensive. Furthermore, the conventional CLIP model doesn't differentiate between the visual semantics and meaning of text regions embedded in images. This can lead to non-robustness when the text in the embedded region doesn't match the image's visual appearance. In this paper, we discuss two effective approaches to improve the efficiency and robustness of CLIP training: (1) augmenting the training dataset while maintaining the same number of optimization steps, and (2) filtering out samples that contain text regions in the image. By doing so, we significantly improve the classification and retrieval accuracy on public benchmarks like ImageNet and CoCo. Filtering out images with text regions also protects the model from typographic attacks. To verify this, we build a new dataset named ImageNet with Adversarial Text Regions (ImageNet-Attr). Our filter-based CLIP model demonstrates a top-1 accuracy of 68.78\%, outperforming previous models whose accuracy was all below 50\%.

arXiv:2305.01860

Towards Imperceptible Document Manipulations against Neural Ranking Models

Adversarial attacks have gained traction in order to identify potential vulnerabilities in neural ranking models (NRMs), but current attack methods often introduce grammatical errors, nonsensical expressions, or incoherent text fragments, which can be easily detected. Additionally, current methods rely heavily on the use of a well-imitated surrogate NRM to guarantee the attack effect, which makes them difficult to use in practice. To address these issues, we propose a framework called Imperceptible DocumEnt Manipulation (IDEM) to produce adversarial documents that are less noticeable to both algorithms and humans. IDEM instructs a well-established generative language model, such as BART, to generate connection sentences without introducing easy-to-detect errors, and employs a separate position-wise merging strategy to balance relevance and coherence of the perturbed text. Experimental results on the popular MS MARCO benchmark demonstrate that IDEM can outperform strong baselines while preserving fluency and correctness of the target documents as evidenced by automatic and human evaluations. Furthermore, the separation of adversarial text generation from the surrogate NRM makes IDEM more robust and less affected by the quality of the surrogate NRM.

arXiv:2303.16378공식 코드 · Papers with Code

A Pilot Study of Query-Free Adversarial Attack against Stable Diffusion

Despite the record-breaking performance in Text-to-Image (T2I) generation by Stable Diffusion, less research attention is paid to its adversarial robustness. In this work, we study the problem of adversarial attack generation for Stable Diffusion and ask if an adversarial text prompt can be obtained even in the absence of end-to-end model queries. We call the resulting problem 'query-free attack generation'. To resolve this problem, we show that the vulnerability of T2I models is rooted in the lack of robustness of text encoders, e.g., the CLIP text encoder used for attacking Stable Diffusion. Based on such insight, we propose both untargeted and targeted query-free attacks, where the former is built on the most influential dimensions in the text embedding space, which we call steerable key dimensions. By leveraging the proposed attacks, we empirically show that only a five-character perturbation to the text prompt is able to cause the significant content shift of synthesized images using Stable Diffusion. Moreover, we show that the proposed target attack can precisely steer the diffusion model to scrub the targeted image content without causing much change in untargeted image content. Our code is available at https://github.com/OPTML-Group/QF-Attack.

arXiv:2303.01234공식 코드 · Papers with Code

Frauds Bargain Attack: Generating Adversarial Text Samples via Word Manipulation Process

Recent research has revealed that natural language processing (NLP) models are vulnerable to adversarial examples. However, the current techniques for generating such examples rely on deterministic heuristic rules, which fail to produce optimal adversarial examples. In response, this study proposes a new method called the Fraud's Bargain Attack (FBA), which uses a randomization mechanism to expand the search space and produce high-quality adversarial examples with a higher probability of success. FBA uses the Metropolis-Hasting sampler, a type of Markov Chain Monte Carlo sampler, to improve the selection of adversarial examples from all candidates generated by a customized stochastic process called the Word Manipulation Process (WMP). The WMP method modifies individual words in a contextually-aware manner through insertion, removal, or substitution. Through extensive experiments, this study demonstrates that FBA outperforms other methods in terms of attack success rate, imperceptibility and sentence quality.

arXiv:2302.11875

Improved Training of Mixture-of-Experts Language GANs

Despite the dramatic success in image generation, Generative Adversarial Networks (GANs) still face great challenges in synthesizing sequences of discrete elements, in particular human language. The difficulty in generator training arises from the limited representation capacity and uninformative learning signals obtained from the discriminator. In this work, we (1) first empirically show that the mixture-of-experts approach is able to enhance the representation capacity of the generator for language GANs and (2) harness the Feature Statistics Alignment (FSA) paradigm to render fine-grained learning signals to advance the generator training. Specifically, FSA forces the mean statistics of the distribution of fake data to approach that of real samples as close as possible in the finite-dimensional feature space. Empirical study on synthetic and real benchmarks shows the superior performance in quantitative evaluation and demonstrates the effectiveness of our approach to adversarial text generation.

arXiv:2302.05892

TextDefense: Adversarial Text Detection based on Word Importance Entropy

Currently, natural language processing (NLP) models are wildly used in various scenarios. However, NLP models, like all deep models, are vulnerable to adversarially generated text. Numerous works have been working on mitigating the vulnerability from adversarial attacks. Nevertheless, there is no comprehensive defense in existing works where each work targets a specific attack category or suffers from the limitation of computation overhead, irresistible to adaptive attack, etc. In this paper, we exhaustively investigate the adversarial attack algorithms in NLP, and our empirical studies have discovered that the attack algorithms mainly disrupt the importance distribution of words in a text. A well-trained model can distinguish subtle importance distribution differences between clean and adversarial texts. Based on this intuition, we propose TextDefense, a new adversarial example detection framework that utilizes the target model's capability to defend against adversarial attacks while requiring no prior knowledge. TextDefense differs from previous approaches, where it utilizes the target model for detection and thus is attack type agnostic. Our extensive experiments show that TextDefense can be applied to different architectures, datasets, and attack methods and outperforms existing methods. We also discover that the leading factor influencing the performance of TextDefense is the target model's generalizability. By analyzing the property of the target model and the property of the adversarial example, we provide our insights into the adversarial attacks in NLP and the principles of our defense method.

arXiv:2302.05120공식 코드 · Papers with Code

Step by Step Loss Goes Very Far: Multi-Step Quantization for Adversarial Text Attacks

We propose a novel gradient-based attack against transformer-based language models that searches for an adversarial example in a continuous space of token probabilities. Our algorithm mitigates the gap between adversarial loss for continuous and discrete text representations by performing multi-step quantization in a quantization-compensation loop. Experiments show that our method significantly outperforms other approaches on various natural language processing (NLP) tasks.

arXiv:2212.11119

A survey on text generation using generative adversarial networks

This work presents a thorough review concerning recent studies and text generation advancements using Generative Adversarial Networks. The usage of adversarial learning for text generation is promising as it provides alternatives to generate the so-called "natural" language. Nevertheless, adversarial text generation is not a simple task as its foremost architecture, the Generative Adversarial Networks, were designed to cope with continuous information (image) instead of discrete data (text). Thus, most works are based on three possible options, i.e., Gumbel-Softmax differentiation, Reinforcement Learning, and modified training objectives. All alternatives are reviewed in this survey as they present the most recent approaches for generating text using adversarial-based techniques. The selected works were taken from renowned databases, such as Science Direct, IEEEXplore, Springer, Association for Computing Machinery, and arXiv, whereas each selected work has been critically analyzed and assessed to present its objective, methodology, and experimental results.

arXiv:2211.09527공식 코드 · Papers with Code

Ignore Previous Prompt: Attack Techniques For Language Models

Transformer-based large language models (LLMs) provide a powerful foundation for natural language tasks in large-scale customer-facing applications. However, studies that explore their vulnerabilities emerging from malicious user interaction are scarce. By proposing PromptInject, a prosaic alignment framework for mask-based iterative adversarial prompt composition, we examine how GPT-3, the most widely deployed language model in production, can be easily misaligned by simple handcrafted inputs. In particular, we investigate two types of attacks -- goal hijacking and prompt leaking -- and demonstrate that even low-aptitude, but sufficiently ill-intentioned agents, can easily exploit GPT-3's stochastic nature, creating long-tail risks. The code for PromptInject is available at https://github.com/agencyenterprise/PromptInject.

arXiv:2210.12813공식 코드 · Papers with Code

TAPE: Assessing Few-shot Russian Language Understanding

Recent advances in zero-shot and few-shot learning have shown promise for a scope of research and practical purposes. However, this fast-growing area lacks standardized evaluation suites for non-English languages, hindering progress outside the Anglo-centric paradigm. To address this line of research, we propose TAPE (Text Attack and Perturbation Evaluation), a novel benchmark that includes six more complex NLU tasks for Russian, covering multi-hop reasoning, ethical concepts, logic and commonsense knowledge. The TAPE's design focuses on systematic zero-shot and few-shot NLU evaluation: (i) linguistic-oriented adversarial attacks and perturbations for analyzing robustness, and (ii) subpopulations for nuanced interpretation. The detailed analysis of testing the autoregressive baselines indicates that simple spelling-based perturbations affect the performance the most, while paraphrasing the input has a more negligible effect. At the same time, the results demonstrate a significant gap between the neural and human baselines for most tasks. We publicly release TAPE (tape-benchmark.com) to foster research on robust LMs that can generalize to new tasks when little to no supervision is available.

arXiv:2206.04137

Adversarial Text Normalization

Text-based adversarial attacks are becoming more commonplace and accessible to general internet users. As these attacks proliferate, the need to address the gap in model robustness becomes imminent. While retraining on adversarial data may increase performance, there remains an additional class of character-level attacks on which these models falter. Additionally, the process to retrain a model is time and resource intensive, creating a need for a lightweight, reusable defense. In this work, we propose the Adversarial Text Normalizer, a novel method that restores baseline performance on attacked content with low computational overhead. We evaluate the efficacy of the normalizer on two problem areas prone to adversarial attacks, i.e. Hate Speech and Natural Language Inference. We find that text normalization provides a task-agnostic defense against character-level attacks that can be implemented supplementary to adversarial retraining solutions, which are more suited for semantic alterations.

arXiv:2205.01287공식 코드 · Papers with Code

SemAttack: Natural Textual Attacks via Different Semantic Spaces

Recent studies show that pre-trained language models (LMs) are vulnerable to textual adversarial attacks. However, existing attack methods either suffer from low attack success rates or fail to search efficiently in the exponentially large perturbation space. We propose an efficient and effective framework SemAttack to generate natural adversarial text by constructing different semantic perturbation functions. In particular, SemAttack optimizes the generated perturbations constrained on generic semantic spaces, including typo space, knowledge space (e.g., WordNet), contextualized semantic space (e.g., the embedding space of BERT clusterings), or the combination of these spaces. Thus, the generated adversarial texts are more semantically close to the original inputs. Extensive experiments reveal that state-of-the-art (SOTA) large-scale LMs (e.g., DeBERTa-v2) and defense strategies (e.g., FreeLB) are still vulnerable to SemAttack. We further demonstrate that SemAttack is general and able to generate natural adversarial texts for different languages (e.g., English and Chinese) with high attack success rates. Human evaluations also confirm that our generated adversarial texts are natural and barely affect human performance. Our code is publicly available at https://github.com/AI-secure/SemAttack.

arXiv:2204.04636공식 코드 · Papers with Code

"That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attacks

Adversarial attacks are a major challenge faced by current machine learning research. These purposely crafted inputs fool even the most advanced models, precluding their deployment in safety-critical applications. Extensive research in computer vision has been carried to develop reliable defense strategies. However, the same issue remains less explored in natural language processing. Our work presents a model-agnostic detector of adversarial text examples. The approach identifies patterns in the logits of the target classifier when perturbing the input text. The proposed detector improves the current state-of-the-art performance in recognizing adversarial inputs and exhibits strong generalization capabilities across different NLP models, datasets, and word-level attacks.

arXiv:2203.07983공식 코드 · Papers with Code

Adversarial Robustness of Neural-Statistical Features in Detection of Generative Transformers

The detection of computer-generated text is an area of rapidly increasing significance as nascent generative models allow for efficient creation of compelling human-like text, which may be abused for the purposes of spam, disinformation, phishing, or online influence campaigns. Past work has studied detection of current state-of-the-art models, but despite a developing threat landscape, there has been minimal analysis of the robustness of detection methods to adversarial attacks. To this end, we evaluate neural and non-neural approaches on their ability to detect computer-generated text, their robustness against text adversarial attacks, and the impact that successful adversarial attacks have on human judgement of text quality. We find that while statistical features underperform neural features, statistical features provide additional adversarial robustness that can be leveraged in ensemble detection models. In the process, we find that previously effective complex phrasal features for detection of computer-generated text hold little predictive power against contemporary generative models, and identify promising statistical features to use instead. Finally, we pioneer the usage of $\Delta$MAUVE as a proxy measure for human judgement of adversarial text quality.

arXiv:2202.09483

Data-Driven Mitigation of Adversarial Text Perturbation

Social networks have become an indispensable part of our lives, with billions of people producing ever-increasing amounts of text. At such scales, content policies and their enforcement become paramount. To automate moderation, questionable content is detected by Natural Language Processing (NLP) classifiers. However, high-performance classifiers are hampered by misspellings and adversarial text perturbations. In this paper, we classify intentional and unintentional adversarial text perturbation into ten types and propose a deobfuscation pipeline to make NLP models robust to such perturbations. We propose Continuous Word2Vec (CW2V), our data-driven method to learn word embeddings that ensures that perturbations of words have embeddings similar to those of the original words. We show that CW2V embeddings are generally more robust to text perturbations than embeddings based on character ngrams. Our robust classification pipeline combines deobfuscation and classification, using proposed defense methods and word embeddings to classify whether Facebook posts are requesting engagement such as likes. Our pipeline results in engagement bait classification that goes from 0.70 to 0.67 AUC with adversarial text perturbation, while character ngram-based word embedding methods result in downstream classification that goes from 0.76 to 0.64.

arXiv:2201.08555

Identifying Adversarial Attacks on Text Classifiers

The landscape of adversarial attacks against text classifiers continues to grow, with new attacks developed every year and many of them available in standard toolkits, such as TextAttack and OpenAttack. In response, there is a growing body of work on robust learning, which reduces vulnerability to these attacks, though sometimes at a high cost in compute time or accuracy. In this paper, we take an alternate approach -- we attempt to understand the attacker by analyzing adversarial text to determine which methods were used to create it. Our first contribution is an extensive dataset for attack detection and labeling: 1.5~million attack instances, generated by twelve adversarial attacks targeting three classifiers trained on six source datasets for sentiment analysis and abuse detection in English. As our second contribution, we use this dataset to develop and benchmark a number of classifiers for attack identification -- determining if a given text has been adversarially manipulated and by which attack. As a third contribution, we demonstrate the effectiveness of three classes of features for these tasks: text properties, capturing content and presentation of text; language model properties, determining which tokens are more or less probable throughout the input; and target model properties, representing how the text classifier is influenced by the attack, including internal node activations. Overall, this represents a first step towards forensics for adversarial attacks against text classifiers.