Learning robust representations of authorial style is crucial for authorship attribution and AI-generated text detection. However, existing methods often struggle with content-style entanglement, where models learn spurious correlations between authors' writing styles and topics, leading to poor generalization across domains. To address this challenge, we propose Explainable Authorship Variational Autoencoder (EAVAE), a novel framework that explicitly disentangles style from content through architectural separation-by-design. EAVAE first pretrains style encoders using supervised contrastive learning on diverse authorship data, then finetunes with a Variational Autoencoder (VEA) architecture using separate encoders for style and content representations. Disentanglement is enforced through a novel discriminator that not only distinguishes whether pairs of style/content representations belong to the same or different authors/content sources, but also generates natural language explanation for their decision, simultaneously mitigating confounding information and enhancing interpretability. Extensive experiments demonstrate the effectiveness of EAVAE. On authorship attribution, we achieve state-of-the-art performance on various datasets, including Amazon Reviews, PAN21, and HRS. For AI-generated text detection, EAVAE excels in few-shot learning over the M4 dataset. Code and data repositories are available online\footnote{https://github.com/hieum98/avae} \footnote{https://huggingface.co/collections/Hieuman/document-level-authorship-datasets}.
National University of Singapore（新加坡国立大学）；AWS AI Research & Education（亚马逊云科技人工智能研究与教育部门）
Authorship attribution is the task of identifying the author of a given text. The key is finding representations that can differentiate between authors. Existing approaches typically use manually designed features that capture a dataset's content and style, but these approaches are dataset-dependent and yield inconsistent performance across corpora. In this work, we propose \textit{learning} author-specific representations by fine-tuning pre-trained generic language representations with a contrastive objective (Contra-X). We show that Contra-X learns representations that form highly separable …
Authorship attribution aims to identify the author of a text based on the stylometric analysis. Authorship obfuscation, on the other hand, aims to protect against authorship attribution by modifying a text's style. In this paper, we evaluate the stealthiness of state-of-the-art authorship obfuscation methods under an adversarial threat model. An obfuscator is stealthy to the extent an adversary finds it challenging to detect whether or not a text modified by the obfuscator is obfuscated - a decision that is key to the adversary interested in authorship attribution. We show that the existing au…
University of California, Davis（加州大学戴维斯分校）；University of Iowa（爱荷华大学）
Recent advances in natural language processing have enabled powerful privacy-invasive authorship attribution. To counter authorship attribution, researchers have proposed a variety of rule-based and learning-based text obfuscation approaches. However, existing authorship obfuscation approaches do not consider the adversarial threat model. Specifically, they are not evaluated against adversarially trained authorship attributors that are aware of potential obfuscation. To fill this gap, we investigate the problem of adversarial authorship attribution for deobfuscation. We show that adversarially…
Penn State University（宾夕法尼亚州立大学）；University of Mississippi（密西西比大学）
Two interlocking research questions of growing interest and importance in privacy research are Authorship Attribution (AA) and Authorship Obfuscation (AO). Given an artifact, especially a text t in question, an AA solution aims to accurately attribute t to its true author out of many candidate authors while an AO solution aims to modify t to hide its true authorship. Traditionally, the notion of authorship and its accompanying privacy concern is only toward human authors. However, in recent years, due to the explosive advancements in Neural Text Generation (NTG) techniques in NLP, capable of s…
It is well known that, within the Latin production of written text, peculiar metric schemes were followed not only in poetic compositions, but also in many prose works. Such metric patterns were based on so-called syllabic quantity, i.e., on the length of the involved syllables, and there is substantial evidence suggesting that certain authors had a preference for certain metric patterns over others. In this research we investigate the possibility to employ syllabic quantity as a base for deriving rhythmic features for the task of computational authorship attribution of Latin prose texts. We t…
Authorship Attribution is a long-standing problem in Natural Language Processing. Several statistical and computational methods have been used to find a solution to this problem. In this paper, we have proposed methods to deal with the authorship attribution problem in Bengali.
With such increasing popularity and availability of digital text data, authorships of digital texts can not be taken for granted due to the ease of copying and parsing. This paper presents a new text style analysis called natural frequency zoned word distribution analysis (NFZ-WDA), and then a basic authorship attribution scheme and an open authorship attribution scheme for digital texts based on the analysis. NFZ-WDA is based on the observation that all authors leave distinct intrinsic word usage traces on texts written by them and these intrinsic styles can be identified and employed to anal…
In this paper, we explore a set of novel features for authorship attribution of documents. These features are derived from a word network representation of natural language text. As has been noted in previous studies, natural language tends to show complex network structure at word level, with low degrees of separation and scale-free (power law) degree distribution. There has also been work on authorship attribution that incorporates ideas from complex networks. The goal of our paper is to explore properties of these complex networks that are suitable as features for machine-learning-based aut…
Institute of Polish Language, Polish Academy of Sciences（波兰科学院波兰语研究所）
In this paper, I introduce a simple method of computing relative word frequencies for authorship attribution and similar stylometric tasks. Rather than computing relative frequencies as the number of occurrences of a given word divided by the total number of tokens in a text, I argue that a more efficient normalization factor is the total number of relevant tokens only. The notion of relevant words includes synonyms and, usually, a few dozen other words in some ways semantically similar to a word in question. To determine such a semantic background, one of word embedding models can be used. Th…
Insight Centre for Data Analytics（数据洞察中心）；Aylien Ltd.（Aylien 有限公司）；National University of Ireland, Galway（爱尔兰国立大学戈尔韦分校）
Convolutional neural networks (CNNs) have demonstrated superior capability for extracting information from raw signals in computer vision. Recently, character-level and multi-channel CNNs have exhibited excellent performance for sentence classification tasks. We apply CNNs to large-scale authorship attribution, which aims to determine an unknown text's author among many candidate authors, motivated by their ability to process character-level signals and to differentiate between a large number of classes, while making fast predictions in comparison to state-of-the-art approaches. We extensively…
Cochin University（科钦大学）；Adi Shankara Institute of Engineering and Technology（阿迪·商羯罗工程技术学院）
Authorship attribution mainly deals with undecided authorship of literary texts. Authorship attribution is useful in resolving issues like uncertain authorship, recognize authorship of unknown texts, spot plagiarism so on. Statistical methods can be used to set apart the approach of an author numerically. The basic methodologies that are made use in computational stylometry are word length, sentence length, vocabulary affluence, frequencies etc. Each author has an inborn style of writing, which is particular to himself. Statistical quantitative techniques can be used to differentiate the appro…
Writing style is a combination of consistent decisions associated with a specific author at different levels of language production, including lexical, syntactic, and structural. In this paper, we introduce a style-aware neural model to encode document information from three stylistic levels and evaluate it in the domain of authorship attribution. First, we propose a simple way to jointly encode syntactic and lexical representations of sentences. Subsequently, we employ an attention-based hierarchical neural network to encode the syntactic and semantic structure of sentences in documents while…
Shahjalal University of Science and Technology（沙贾拉勒科技大学）
Characters are the smallest unit of text that can extract stylometric signals to determine the author of a text. In this paper, we investigate the effectiveness of character-level signals in Authorship Attribution of Bangla Literature and show that the results are promising but improvable. The time and memory efficiency of the proposed model is much higher than the word level counterparts but accuracy is 2-5% less than the best performing word-level models. Comparison of various word-based models is performed and shown that the proposed model performs increasingly better with larger datasets. …
Cross-language authorship attribution problems rely on either translation to enable the use of single-language features, or language-independent feature extraction methods. Until recently, the lack of datasets for this problem hindered the development of the latter, and single-language solutions were performed on machine-translated corpora. In this paper, we present a novel language-independent feature for authorship analysis based on dependency graphs and universal part of speech tags, called DT-grams (dependency tree grams), which are constructed by selecting specific sub-parts of the depend…
Do bilingual Russian-French authors of the end of the twentieth century such as Andreï Makine, Valéry Afanassiev, Vladimir Fédorovski, Iegor Gran, Luba Jurgenson have common stylistic traits in the novels they wrote in French? Can we distinguish between them and non-bilingual French writers' texts? Is the phenomenon of interference observable in French texts of Russian authors? This paper applies authorship attribution methods including Support Vector Machine (SVM), $K$-Nearest Neighbors (KNN), Ridge classification, and Neural Network to answer these questions.
Wolfram Research（沃尔夫勒姆研究公司）；Faculty of Sciences, University of Craiova（克拉约瓦大学理学院）
The Chaos Game Representation, a method for creating images from nucleotide sequences, is modified to make images from chunks of text documents. Machine learning methods are then applied to train classifiers based on authorship. Experiments are conducted on several benchmark data sets in English, including the widely used Federalist Papers, and one in Portuguese. Validation results for the trained classifiers are competitive with the best methods in prior literature. The methodology is also successfully applied for text categorization with encouraging results. One classifier method is moreover…
We explore techniques to maximize the effectiveness of discourse information in the task of authorship attribution. We present a novel method to embed discourse features in a Convolutional Neural Network text classifier, which achieves a state-of-the-art result by a substantial margin. We empirically investigate several featurization methods to understand the conditions under which discourse features contribute non-trivial performance gains, and analyze discourse embeddings.
Institute of Linguistics and Language Technology, University of Malta（马耳他大学语言学与语言技术研究所）；Center for Language and Cognition, University of Groningen（格罗宁根大学语言与认知中心）
Existing research on Authorship Attribution (AA) focuses on texts for which a lot of data is available (e.g novels), mainly in English. We approach AA via Authorship Verification on short Italian texts in two novel datasets, and analyze the interaction between genre, topic, gender and length. Results show that AV is feasible even with little data, but more evidence helps. Gender and topic can be indicative clues, and if not controlled for, they might overtake more specific aspects of personal style.
DEVCOM Army Research Laboratory（DEVCOM陆军研究实验室）；Carnegie Mellon University（卡内基梅隆大学）；Duke University（杜克大学）
Despite decades of research on authorship attribution (AA) and authorship verification (AV), inconsistent dataset splits/filtering and mismatched evaluation methods make it difficult to assess the state of the art. In this paper, we present a survey of the fields, resolve points of confusion, introduce Valla that standardizes and benchmarks AA/AV datasets and metrics, provide a large-scale empirical evaluation, and provide apples-to-apples comparisons between existing methods. We evaluate eight promising methods on fifteen datasets (including distribution-shifted challenge sets) and introduce …
A method for authorship attribution based on function word adjacency networks (WANs) is introduced. Function words are parts of speech that express grammatical relationships between other words but do not carry lexical meaning on their own. In the WANs in this paper, nodes are function words and directed edges stand in for the likelihood of finding the sink word in the ordered vicinity of the source word. WANs of different authors can be interpreted as transition probabilities of a Markov chain and are therefore compared in terms of their relative entropies. Optimal selection of WAN parameters…
We describe a technique for attributing parts of a written text to a set of unknown authors. Nothing is assumed to be known a priori about the writing styles of potential authors. We use multiple independent clusterings of an input text to identify parts that are similar and dissimilar to one another. We describe algorithms necessary to combine the multiple clusterings into a meaningful output. We show results of the application of the technique on texts having multiple writing styles.
Writing style is a combination of consistent decisions at different levels of language production including lexical, syntactic, and structural associated to a specific author (or author groups). While lexical-based models have been widely explored in style-based text classification, relying on content makes the model less scalable when dealing with heterogeneous data comprised of various topics. On the other hand, syntactic models which are content-independent, are more robust against topic variance. In this paper, we introduce a syntactic recurrent neural network to encode the syntactic patte…
Existing research in computational authorship attribution (AA) has primarily focused on attribution tasks with a limited number of authors in a closed-set configuration. This restricted set-up is far from being realistic in dealing with highly entangled real-world AA tasks that involve a large number of candidate authors for attribution during test time. In this paper, we study AA in historical texts using anew data set compiled from the Victorian literature. We investigate the predictive capacity of most common English words in distinguishing writings of most prominent Victorian novelists. We…
Authorship attribution is the problem of identifying the most plausible author of an anonymous text from a set of candidate authors. Researchers have investigated same-topic and cross-topic scenarios of authorship attribution, which differ according to whether new, unseen topics are used in the testing phase. However, neither scenario allows us to explain whether errors are caused by a failure to capture authorship writing style or by a topic shift. Motivated by this, we propose the \emph{topic confusion} task where we switch the author-topic configuration between the training and testing sets…
Ruhr University Bochum（波鸿鲁尔大学）；Bucknell University（巴克内尔大学）
Traditional computational authorship attribution describes a classification task in a closed-set scenario. Given a finite set of candidate authors and corresponding labeled texts, the objective is to determine which of the authors has written another set of anonymous or disputed texts. In this work, we propose a probabilistic autoencoding framework to deal with this supervised classification task. More precisely, we are extending a variational autoencoder (VAE) with embedded Gaussian mixture model to a Student-$t$ mixture model. Autoencoders have had tremendous success in learning latent repre…
University of Antwerp（安特卫普大学）；Ghent University（根特大学）
Recent applications of neural language models have led to an increased interest in the automatic generation of natural language. However impressive, the evaluation of neurally generated text has so far remained rather informal and anecdotal. Here, we present an attempt at the systematic assessment of one aspect of the quality of neurally generated text. We focus on a specific aspect of neural language generation: its ability to reproduce authorial writing styles. Using established models for authorship attribution, we empirically assess the stylistic qualities of neurally generated text. In co…
University of Toronto（多伦多大学）；University of São Paulo（圣保罗大学）
Concepts and methods of complex networks can be used to analyse texts at their different complexity levels. Examples of natural language processing (NLP) tasks studied via topological analysis of networks are keyword identification, automatic extractive summarization and authorship attribution. Even though a myriad of network measurements have been applied to study the authorship attribution problem, the use of motifs for text analysis has been restricted to a few works. The goal of this paper is to apply the concept of motifs, recurrent interconnection patterns, in the authorship attribution …
In practice, training language models for individual authors is often expensive because of limited data resources. In such cases, Neural Network Language Models (NNLMs), generally outperform the traditional non-parametric N-gram models. Here we investigate the performance of a feed-forward NNLM on an authorship attribution problem, with moderate author set size and relatively limited data. We also consider how the text topics impact performance. Compared with a well-constructed N-gram baseline method with Kneser-Ney smoothing, the proposed method achieves nearly 2:5% reduction in perplexity an…
Syntactic structure of sentences in a document substantially informs about its authorial writing style. Sentence representation learning has been widely explored in recent years and it has been shown that it improves the generalization of different downstream tasks across many domains. Even though utilizing probing methods in several studies suggests that these learned contextual representations implicitly encode some amount of syntax, explicit syntactic information further improves the performance of deep neural models in the domain of authorship attribution. These observations have motivated…
São Carlos Institute of Physics, University of São Paulo（圣保罗大学圣保罗物理研究所）；Institute of Mathematics and Computer Science, University of São Paulo（圣保罗大学数学与计算机科学研究所）
The authorship attribution is a problem of considerable practical and technical interest. Several methods have been designed to infer the authorship of disputed documents in multiple contexts. While traditional statistical methods based solely on word counts and related measurements have provided a simple, yet effective solution in particular cases; they are prone to manipulation. Recently, texts have been successfully modeled as networks, where words are represented by nodes linked according to textual similarity measurements. Such models are useful to identify informative topological pattern…
Recently, there has been a rise in the development of powerful pre-trained natural language models, including GPT-2, Grover, and XLM. These models have shown state-of-the-art capabilities towards a variety of different NLP tasks, including question answering, content summarisation, and text generation. Alongside this, there have been many studies focused on online authorship attribution (AA). That is, the use of models to identify the authors of online texts. Given the power of natural language models in generating convincing texts, this paper examines the degree to which these language models…
University of Electronic Science and Technology of China（电子科技大学）；Singapore Management University（新加坡管理大学）；University of Saskatchewan（萨斯喀彻温大学）
Authorship attribution (AA), which is the task of finding the owner of a given text, is an important and widely studied research topic with many applications. Recent works have shown that deep learning methods could achieve significant accuracy improvement for the AA task. Nevertheless, most of these proposed methods represent user posts using a single type of feature (e.g., word bi-grams) and adopt a text classification approach to address the task. Furthermore, these methods offer very limited explainability of the AA results. In this paper, we address these limitations by proposing DeepStyl…
Microsoft Corporation（微软公司）；Guru Gobind Singh Indraprastha University（古鲁·戈宾德·辛格因德拉普拉斯塔大学）
The writing style of a person can be affirmed as a unique identity indicator; the words used, and the structuring of the sentences are clear measures which can identify the author of a specific work. Stylometry and its subset - Authorship Attribution, have a long history beginning from the 19th century, and we can still find their use in modern times. The emergence of the Internet has shifted the application of attribution studies towards non-standard texts that are comparatively shorter to and different from the long texts on which most research has been done. The aim of this paper focuses on…
Well-established automatic analyses of texts mainly consider frequencies of linguistic units, e.g. letters, words and bigrams, while methods based on co-occurrence networks consider the structure of texts regardless of the nodes label (i.e. the words semantics). In this paper, we reconcile these distinct viewpoints by introducing a generalized similarity measure to compare texts which accounts for both the network structure of texts and the role of individual words in the networks. We use the similarity measure for authorship attribution of three collections of books, each composed of 8 author…
