Recent advancements in large language models (LLMs) have been fueled by large scale training corpora drawn from diverse sources such as websites, news articles, and books. These datasets often contain explicit user information, such as person names and addresses, that LLMs may unintentionally reproduce in their generated outputs. Beyond such explicit content, LLMs can also leak identity revealing cues through implicit signals such as distinctive writing styles, raising significant concerns about authorship privacy. There are three major automated tasks in authorship privacy, namely authorship obfuscation (AO), authorship mimicking (AM), and authorship verification (AV). Prior research has studied AO, AM, and AV independently. However, their interplays remain under explored, which leaves a major research gap, especially in the era of LLMs, where they are profoundly shaping how we curate and share user generated content, and the distinction between machine generated and human authored text is also increasingly blurred. This work then presents the first unified framework for analyzing the dynamic relationships among LLM enabled AO, AM, and AV in the context of authorship privacy. We quantify how they interact with each other to transform human authored text, examining effects at a single point in time and iteratively over time. We also examine the role of demographic metadata, such as gender, academic background, in modulating their performances, inter-task dynamics, and privacy risks. All source code will be publicly available.
As human-AI collaboration becomes increasingly prevalent in educational contexts, understanding and measuring the extent and nature of such interactions pose significant challenges. This research investigates the use of authorship verification (AV) techniques not as a punitive measure, but as a means to quantify AI assistance in academic writing, with a focus on promoting transparency, interpretability, and student development. Building on prior work, we structured our investigation into three stages: dataset selection and expansion, AV method development, and systematic evaluation. Using three datasets - including a public dataset (PAN-14) and two from University of Melbourne students from various courses - we expanded the data to include LLM-generated texts, totalling 1,889 documents and 540 authorship problems from 506 students. We developed an adapted Feature Vector Difference AV methodology to construct robust academic writing profiles for students, designed to capture meaningful, individual characteristics of their writing. The method's effectiveness was evaluated across multiple scenarios, including distinguishing between student-authored and LLM-generated texts and testing resilience against LLMs' attempts to mimic student writing styles. Results demonstrate the enhanced AV classifier's ability to identify stylometric discrepancies and measure human-AI collaboration at word and sentence levels while providing educators with a transparent tool to support academic integrity investigations. This work advances AV technology, offering actionable insights into the dynamics of academic writing in an AI-driven era.
The increasing use of Artificial Intelligence (AI) technologies, such as Large Language Models (LLMs) has led to nontrivial improvements in various tasks, including accurate authorship identification of documents. However, while LLMs improve such defense techniques, they also simultaneously provide a vehicle for malicious actors to launch new attack vectors. To combat this security risk, we evaluate the adversarial robustness of authorship models (specifically an authorship verification model) to potent LLM-based attacks. These attacks include untargeted methods - \textit{authorship obfuscation} and targeted methods - \textit{authorship impersonation}. For both attacks, the objective is to mask or mimic the writing style of an author while preserving the original texts' semantics, respectively. Thus, we perturb an accurate authorship verification model, and achieve maximum attack success rates of 92\% and 78\% for both obfuscation and impersonation attacks, respectively.
Variation in language is ubiquitous and often systematically linked to regional, social, and contextual factors. Tokenizers split texts into smaller units and might behave differently for less common linguistic forms. This might affect downstream LLM performance differently on two types of tasks: Tasks where the model should be robust to language variation (e.g., for semantic tasks like NLI, labels do not depend on whether a text uses British or American spelling) and tasks where the model should be sensitive to language variation (e.g., for form-based tasks like authorship verification, labels depend on whether a text uses British or American spelling). We pre-train BERT base models for the popular Byte-Pair Encoding algorithm to investigate how key algorithmic design choices impact downstream models' performances: fitting corpus, pre-tokenizer and vocabulary size. We find that the best tokenizer varies on the two task types -- with the pre-tokenizer having the biggest impact on performance. Further, we introduce a new approach to estimate tokenizer impact on downstream LLM performance, showing significant improvement over techniques like R\'enyi efficiency. We encourage more work on language variation and its relation to tokenizers and thus LLM performance.
The Questio de aqua et terra is a cosmological treatise traditionally attributed to Dante Alighieri. However, the authenticity of this text is controversial, due to discrepancies with Dante's established works and to the absence of contemporary references. This study investigates the authenticity of the Questio via computational authorship verification (AV), a class of techniques which combine supervised machine learning and stylometry. We build a family of AV systems and assemble a corpus of 330 13th- and 14th-century Latin texts, which we use to comparatively evaluate the AV systems through leave-one-out cross-validation. Our best-performing system achieves high verification accuracy (F1=0.970) despite the heterogeneity of the corpus in terms of textual genre. The key contribution to the accuracy of this system is shown to come from Distributional Random Oversampling (DRO), a technique specially tailored to text classification which is here used for the first time in AV. The application of the AV system to the Questio returns a highly confident prediction concerning its authenticity. These findings contribute to the debate on the authorship of the Questio, and highlight DRO's potential in the application of AV to cultural heritage.
This paper evaluates the performance of Large Language Models (LLMs) in authorship attribution and authorship verification tasks for Latin texts of the Patristic Era. The study showcases that LLMs can be robust in zero-shot authorship verification even on short texts without sophisticated feature engineering. Yet, the models can also be easily "mislead" by semantics. The experiments also demonstrate that steering the model's authorship analysis and decision-making is challenging, unlike what is reported in the studies dealing with high-resource modern languages. Although LLMs prove to be able to beat, under certain circumstances, the traditional baselines, obtaining a nuanced and truly explainable decision requires at best a lot of experimentation.
Authorship verification (AV) aims to identify whether a pair of texts has the same author. We address the challenge of evaluating AV models' robustness against topic shifts. The conventional evaluation assumes minimal topic overlap between training and test data. However, we argue that there can still be topic leakage in test data, causing misleading model performance and unstable rankings. To address this, we propose an evaluation method called Heterogeneity-Informed Topic Sampling (HITS), which creates a smaller dataset with a heterogeneously distributed topic set. Our experimental results demonstrate that HITS-sampled datasets yield a more stable ranking of models across random seeds and evaluation splits. Our contributions include: 1. An analysis of causes and effects of topic leakage. 2. A demonstration of the HITS in reducing the effects of topic leakage, and 3. The Robust Authorship Verification bENchmark (RAVEN) that allows topic shortcut test to uncover AV models' reliance on topic-specific features.
Large Language Models (LLMs) have demonstrated remarkable proficiency in a wide range of NLP tasks. However, when it comes to authorship verification (AV) tasks, which involve determining whether two given texts share the same authorship, even advanced models like ChatGPT exhibit notable limitations. This paper introduces a novel approach, termed InstructAV, for authorship verification. This approach utilizes LLMs in conjunction with a parameter-efficient fine-tuning (PEFT) method to simultaneously improve accuracy and explainability. The distinctiveness of InstructAV lies in its ability to align classification decisions with transparent and understandable explanations, representing a significant progression in the field of authorship verification. Through comprehensive experiments conducted across various datasets, InstructAV demonstrates its state-of-the-art performance on the AV task, offering high classification accuracy coupled with enhanced explanation reliability.
Authorship Verification (AV) (do two documents have the same author?) is essential in many sensitive real-life applications. AV is often used in proprietary domains that require a private, offline model, making SOTA online models like ChatGPT undesirable. Current offline models however have lower downstream utility due to low accuracy/scalability (eg: traditional stylometry AV systems) and lack of accessible post-hoc explanations. In this work, we take the first step to address the above challenges with our trained, offline Llama-3-8B model CAVE (Controllable Authorship Verification Explanations): CAVE generates free-text AV explanations that are controlled to be (1) structured (can be decomposed into sub-explanations in terms of relevant linguistic features), and (2) easily verified for explanation-label consistency (via intermediate labels in sub-explanations). We first engineer a prompt that can generate silver training data from a SOTA teacher model in the desired CAVE output format. We then filter and distill this data into a pretrained Llama-3-8B, our carefully selected student model. Results on three difficult AV datasets IMDb62, Blog-Auth, and Fanfiction show that CAVE generates high quality explanations (as measured by automatic and human evaluation) as well as competitive task accuracies.
Humans naturally attribute utterances of direct speech to their speaker in literary works. When attributing quotes, we process contextual information but also access mental representations of characters that we build and revise throughout the narrative. Recent methods to automatically attribute such utterances have explored simulating human logic with deterministic rules or learning new implicit rules with neural networks when processing contextual information. However, these systems inherently lack \textit{character} representations, which often leads to errors in more challenging examples of attribution: anaphoric and implicit quotes. In this work, we propose to augment a popular quotation attribution system, BookNLP, with character embeddings that encode global stylistic information of characters derived from an off-the-shelf stylometric model, Universal Authorship Representation (UAR). We create DramaCV (Code and data can be found at https://github.com/deezer/character_embeddings_qa ), a corpus of English drama plays from the 15th to 20th century that we automatically annotate for Authorship Verification of fictional characters utterances, and release two versions of UAR trained on DramaCV, that are tailored for literary characters analysis. Then, through an extensive evaluation on 28 novels, we show that combining BookNLP's contextual information with our proposed global character embeddings improves the identification of speakers for anaphoric and implicit quotes, reaching state-of-the-art performance.
Large Language Models (LLMs) have revolutionised the field of Natural Language Processing (NLP) and have achieved state-of-the-art performance in practically every task in this field. However, the prevalent approach used in text generation, Causal Language Modelling (CLM), which generates text sequentially from left to right, inherently limits the freedom of the model, which does not decide when and where each token is generated. In contrast, Masked Language Modelling (MLM), primarily used for language understanding tasks, can generate tokens anywhere in the text and any order. This paper conducts an extensive comparison of MLM and CLM approaches for text generation tasks. To do so, we pre-train several language models of comparable sizes on three different datasets, namely 1) medical discharge summaries, 2) movie plot synopses, and 3) authorship verification datasets. To assess the quality of the generations, we first employ quantitative metrics and then perform a qualitative human evaluation to analyse coherence and grammatical correctness. In addition, we evaluate the usefulness of the generated texts by using them in three different downstream tasks: 1) Entity Recognition, 2) Text Classification, and 3) Authorship Verification. The results show that MLM consistently outperforms CLM in text generation across all datasets, with higher quantitative scores and better coherence in the generated text. The study also finds \textit{no strong correlation} between the quality of the generated text and the performance of the models in the downstream tasks. With this study, we show that MLM for text generation has great potential for future research and provides direction for future studies in this area.
Authorship Verification (AV) is a text classification task concerned with inferring whether a candidate text has been written by one specific author or by someone else. It has been shown that many AV systems are vulnerable to adversarial attacks, where a malicious author actively tries to fool the classifier by either concealing their writing style, or by imitating the style of another author. In this paper, we investigate the potential benefits of augmenting the classifier training set with (negative) synthetic examples. These synthetic examples are generated to imitate the style of the author of interest. We analyze the improvements in classifier prediction that this augmentation brings to bear in the task of AV in an adversarial setting. In particular, we experiment with three different generator architectures (one based on Recurrent Neural Networks, another based on small-scale transformers, and another based on the popular GPT model) and with two training strategies (one inspired by standard Language Models, and another inspired by Wasserstein Generative Adversarial Networks). We evaluate our hypothesis on five datasets (three of which have been specifically collected to represent an adversarial setting) and using two learning algorithms for the AV classifier (Support Vector Machines and Convolutional Neural Networks). This experimentation has yielded negative results, revealing that, although our methodology proves effective in many adversarial settings, its benefits are too sporadic for a pragmatical application.
Authorship Verification (AV) is the process of analyzing a set of documents to determine whether they were written by a specific author. This problem often arises in forensic scenarios, e.g., in cases where the documents in question constitute evidence for a crime. Existing state-of-the-art AV methods use computational solutions that are not supported by a plausible scientific explanation for their functioning and that are often difficult for analysts to interpret. To address this, we propose a method relying on calculating a quantity we call $\lambda_G$ (LambdaG): the ratio between the likelihood of a document given a model of the Grammar for the candidate author and the likelihood of the same document given a model of the Grammar for a reference population. These Grammar Models are estimated using $n$-gram language models that are trained solely on grammatical features. Despite not needing large amounts of data for training, LambdaG still outperforms other established AV methods with higher computational complexity, including a fine-tuned Siamese Transformer network. Our empirical evaluation based on four baseline methods applied to twelve datasets shows that LambdaG leads to better results in terms of both accuracy and AUC in eleven cases and in all twelve cases if considering only topic-agnostic methods. The algorithm is also highly robust to important variations in the genre of the reference population in many cross-genre comparisons. In addition to these properties, we demonstrate how LambdaG is easier to interpret than the current state-of-the-art. We argue that the advantage of LambdaG over other methods is due to fact that it is compatible with Cognitive Linguistic theories of language processing.
The ability to accurately identify authorship is crucial for verifying content authenticity and mitigating misinformation. Large Language Models (LLMs) have demonstrated an exceptional capacity for reasoning and problem-solving. However, their potential in authorship analysis remains under-explored. Traditional studies have depended on hand-crafted stylistic features, whereas state-of-the-art approaches leverage text embeddings from pre-trained language models. These methods, which typically require fine-tuning on labeled data, often suffer from performance degradation in cross-domain applications and provide limited explainability. This work seeks to address three research questions: (1) Can LLMs perform zero-shot, end-to-end authorship verification effectively? (2) Are LLMs capable of accurately attributing authorship among multiple candidates authors (e.g., 10 and 20)? (3) Can LLMs provide explainability in authorship analysis, particularly through the role of linguistic features? Moreover, we investigate the integration of explicit linguistic features to guide LLMs in their reasoning processes. Our assessment demonstrates LLMs' proficiency in both tasks without the need for domain-specific fine-tuning, providing explanations into their decision making via a detailed analysis of linguistic features. This establishes a new benchmark for future research on LLM-based authorship analysis.
Recent approaches to automatically detect the speaker of an utterance of direct speech often disregard general information about characters in favor of local information found in the context, such as surrounding mentions of entities. In this work, we explore stylistic representations of characters built by encoding their quotes with off-the-shelf pretrained Authorship Verification models in a large corpus of English novels (the Project Dialogism Novel Corpus). Results suggest that the combination of stylistic and topical information captured in some of these models accurately distinguish characters among each other, but does not necessarily improve over semantic-only models when attributing quotes. However, these results vary across novels and more investigation of stylometric models particularly tailored for literary texts and the study of characters should be conducted.
Authorship verification is the task of determining if two distinct writing samples share the same author and is typically concerned with the attribution of written text. In this paper, we explore the attribution of transcribed speech, which poses novel challenges. The main challenge is that many stylistic features, such as punctuation and capitalization, are not informative in this setting. On the other hand, transcribed speech exhibits other patterns, such as filler words and backchannels (e.g., 'um', 'uh-huh'), which may be characteristic of different speakers. We propose a new benchmark for speaker attribution focused on human-transcribed conversational speech transcripts. To limit spurious associations of speakers with topic, we employ both conversation prompts and speakers participating in the same conversation to construct verification trials of varying difficulties. We establish the state of the art on this new benchmark by comparing a suite of neural and non-neural baselines, finding that although written text attribution models achieve surprisingly good performance in certain settings, they perform markedly worse as conversational topic is increasingly controlled. We present analyses of the impact of transcription style on performance as well as the ability of fine-tuning on speech transcripts to improve performance.
While a substantial amount of work has recently been devoted to enhance the performance of computational Authorship Identification (AId) systems, little to no attention has been paid to endowing AId systems with the ability to explain the reasons behind their predictions. This lacking substantially hinders the practical employment of AId methodologies, since the predictions returned by such systems are hardly useful unless they are supported with suitable explanations. In this paper, we explore the applicability of existing general-purpose eXplainable Artificial Intelligence (XAI) techniques to AId, with a special focus on explanations addressed to scholars working in cultural heritage. In particular, we assess the relative merits of three different types of XAI techniques (feature ranking, probing, factuals and counterfactual selection) on three different AId tasks (authorship attribution, authorship verification, same-authorship verification) by running experiments on real AId data. Our analysis shows that, while these techniques make important first steps towards explainable Authorship Identification, more work remains to be done in order to provide tools that can be profitably integrated in the workflows of scholars.
Authorship verification (AV) is a fundamental task in natural language processing (NLP) and computational linguistics, with applications in forensic analysis, plagiarism detection, and identification of deceptive content. Existing AV techniques, including traditional stylometric and deep learning approaches, face limitations in terms of data requirements and lack of explainability. To address these limitations, this paper proposes PromptAV, a novel technique that leverages Large-Language Models (LLMs) for AV by providing step-by-step stylometric explanation prompts. PromptAV outperforms state-of-the-art baselines, operates effectively with limited training data, and enhances interpretability through intuitive explanations, showcasing its potential as an effective and interpretable solution for the AV task.
Human trafficking (HT) is a pervasive global issue affecting vulnerable individuals, violating their fundamental human rights. Investigations reveal that a significant number of HT cases are associated with online advertisements (ads), particularly in escort markets. Consequently, identifying and connecting HT vendors has become increasingly challenging for Law Enforcement Agencies (LEAs). To address this issue, we introduce IDTraffickers, an extensive dataset consisting of 87,595 text ads and 5,244 vendor labels to enable the verification and identification of potential HT vendors on online escort markets. To establish a benchmark for authorship identification, we train a DeCLUTR-small model, achieving a macro-F1 score of 0.8656 in a closed-set classification environment. Next, we leverage the style representations extracted from the trained classifier to conduct authorship verification, resulting in a mean r-precision score of 0.8852 in an open-set ranking environment. Finally, to encourage further research and ensure responsible data sharing, we plan to release IDTraffickers for the authorship attribution task to researchers under specific conditions, considering the sensitive nature of the data. We believe that the availability of our dataset and benchmarks will empower future researchers to utilize our findings, thereby facilitating the effective linkage of escort ads and the development of more robust approaches for identifying HT indicators.
While state-of-the-art large language models (LLMs) can excel at adapting text from one style to another, current work does not address the explainability of style transfer models. Recent work has explored generating textual explanations from larger teacher models and distilling them into smaller student models. One challenge with such approach is that LLM outputs may contain errors that require expertise to correct, but gathering and incorporating expert feedback is difficult due to cost and availability. To address this challenge, we propose ICLEF, a novel human-AI collaboration approach to model distillation that incorporates scarce expert human feedback by combining in-context learning and model self-critique. We show that our method leads to generation of high-quality synthetic explainable style transfer datasets for formality (e-GYAFC) and subjective bias (e-WNC). Via automatic and human evaluation, we show that specialized student models fine-tuned on our datasets outperform generalist teacher models on the explainable style transfer task in one-shot settings, and perform competitively compared to few-shot teacher models, highlighting the quality of the data and the role of expert feedback. In an extrinsic task of authorship attribution, we show that explanations generated by smaller models fine-tuned on e-GYAFC are more predictive of authorship than explanations generated by few-shot teacher models.