arXiv subjects

Stephen Meisenbacher

Publications and source records attributed to Stephen Meisenbacher.

32 records · Page 2Linked to original sources

arXiv2025

On the Impact of Noise in Differentially Private Text Rewriting

The field of text privatization often leverages the notion of $\textit{Differential Privacy}$ (DP) to provide formal guarantees in the rewriting or obfuscation of sensitive textual data. A common and nearly ubiquitous form of DP application necessitates the addition of calibrated noise to vector representations of text, either at the data- or model-level, which is governed by the privacy parameter $\varepsilon$. However, noise addition almost undoubtedly leads to considerable utility loss, thereby highlighting one major drawback of DP in NLP. In this work, we introduce a new sentence infilling privatization technique, and we use this method to explore the effect of noise in DP text rewriting. We empirically demonstrate that non-DP privatization techniques excel in utility preservation and can find an acceptable empirical privacy-utility trade-off, yet cannot outperform DP methods in empirical privacy protections. Our results highlight the significant impact of noise in current DP rewriting mechanisms, leading to a discussion of the merits and challenges of DP in NLP, as well as the opportunities that non-DP methods present.

cs.CL↗

arXiv2024

Thinking Outside of the Differential Privacy Box: A Case Study in Text Privatization with Language Model Prompting

The field of privacy-preserving Natural Language Processing has risen in popularity, particularly at a time when concerns about privacy grow with the proliferation of Large Language Models. One solution consistently appearing in recent literature has been the integration of Differential Privacy (DP) into NLP techniques. In this paper, we take these approaches into critical view, discussing the restrictions that DP integration imposes, as well as bring to light the challenges that such restrictions entail. To accomplish this, we focus on $\textbf{DP-Prompt}$, a recent method for text privatization leveraging language models to rewrite texts. In particular, we explore this rewriting task in multiple scenarios, both with DP and without DP. To drive the discussion on the merits of DP in NLP, we conduct empirical utility and privacy experiments. Our results demonstrate the need for more discussion on the usability of DP in NLP and its benefits over non-DP approaches.

cs.CL↗

arXiv2024

An Improved Method for Class-specific Keyword Extraction: A Case Study in the German Business Registry

The task of $\textit{keyword extraction}$ is often an important initial step in unsupervised information extraction, forming the basis for tasks such as topic modeling or document classification. While recent methods have proven to be quite effective in the extraction of keywords, the identification of $\textit{class-specific}$ keywords, or only those pertaining to a predefined class, remains challenging. In this work, we propose an improved method for class-specific keyword extraction, which builds upon the popular $\textbf{KeyBERT}$ library to identify only keywords related to a class described by $\textit{seed keywords}$. We test this method using a dataset of German business registry entries, where the goal is to classify each business according to an economic sector. Our results reveal that our method greatly improves upon previous approaches, setting a new standard for $\textit{class-specific}$ keyword extraction.

cs.CL↗

arXiv2024

Privacy Risks of General-Purpose AI Systems: A Foundation for Investigating Practitioner Perspectives

The rise of powerful AI models, more formally $\textit{General-Purpose AI Systems}$ (GPAIS), has led to impressive leaps in performance across a wide range of tasks. At the same time, researchers and practitioners alike have raised a number of privacy concerns, resulting in a wealth of literature covering various privacy risks and vulnerabilities of AI models. Works surveying such risks provide differing focuses, leading to disparate sets of privacy risks with no clear unifying taxonomy. We conduct a systematic review of these survey papers to provide a concise and usable overview of privacy risks in GPAIS, as well as proposed mitigation strategies. The developed privacy framework strives to unify the identified privacy risks and mitigations at a technical level that is accessible to non-experts. This serves as the basis for a practitioner-focused interview study to assess technical stakeholder perceptions of privacy risks and mitigations in GPAIS.

cs.CY↗

arXiv2024

DP-MLM: Differentially Private Text Rewriting Using Masked Language Models

The task of text privatization using Differential Privacy has recently taken the form of $\textit{text rewriting}$, in which an input text is obfuscated via the use of generative (large) language models. While these methods have shown promising results in the ability to preserve privacy, these methods rely on autoregressive models which lack a mechanism to contextualize the private rewriting process. In response to this, we propose $\textbf{DP-MLM}$, a new method for differentially private text rewriting based on leveraging masked language models (MLMs) to rewrite text in a semantically similar $\textit{and}$ obfuscated manner. We accomplish this with a simple contextualization technique, whereby we rewrite a text one token at a time. We find that utilizing encoder-only MLMs provides better utility preservation at lower $\varepsilon$ levels, as compared to previous methods relying on larger models with a decoder. In addition, MLMs allow for greater customization of the rewriting mechanism, as opposed to generative approaches. We make the code for $\textbf{DP-MLM}$ public and reusable, found at https://github.com/sjmeis/DPMLM .

cs.CL↗

arXiv2024

A Collocation-based Method for Addressing Challenges in Word-level Metric Differential Privacy

Applications of Differential Privacy (DP) in NLP must distinguish between the syntactic level on which a proposed mechanism operates, often taking the form of $\textit{word-level}$ or $\textit{document-level}$ privatization. Recently, several word-level $\textit{Metric}$ Differential Privacy approaches have been proposed, which rely on this generalized DP notion for operating in word embedding spaces. These approaches, however, often fail to produce semantically coherent textual outputs, and their application at the sentence- or document-level is only possible by a basic composition of word perturbations. In this work, we strive to address these challenges by operating $\textit{between}$ the word and sentence levels, namely with $\textit{collocations}$. By perturbing n-grams rather than single words, we devise a method where composed privatized outputs have higher semantic coherence and variable length. This is accomplished by constructing an embedding model based on frequently occurring word groups, in which unigram words co-exist with bi- and trigram collocations. We evaluate our method in utility and privacy tests, which make a clear case for tokenization strategies beyond the word level.

cs.CL↗

arXiv2024

Just Rewrite It Again: A Post-Processing Method for Enhanced Semantic Similarity and Privacy Preservation of Differentially Private Rewritten Text

The study of Differential Privacy (DP) in Natural Language Processing often views the task of text privatization as a $\textit{rewriting}$ task, in which sensitive input texts are rewritten to hide explicit or implicit private information. In order to evaluate the privacy-preserving capabilities of a DP text rewriting mechanism, $\textit{empirical privacy}$ tests are frequently employed. In these tests, an adversary is modeled, who aims to infer sensitive information (e.g., gender) about the author behind a (privatized) text. Looking to improve the empirical protections provided by DP rewriting methods, we propose a simple post-processing method based on the goal of aligning rewritten texts with their original counterparts, where DP rewritten texts are rewritten $\textit{again}$. Our results show that such an approach not only produces outputs that are more semantically reminiscent of the original inputs, but also texts which score on average better in empirical privacy evaluations. Therefore, our approach raises the bar for DP rewriting methods in their empirical privacy evaluations, providing an extra layer of protection against malicious adversaries.

cs.CL↗

arXiv2024

Towards A Structured Overview of Use Cases for Natural Language Processing in the Legal Domain: A German Perspective

In recent years, the field of Legal Tech has risen in prevalence, as the Natural Language Processing (NLP) and legal disciplines have combined forces to digitalize legal processes. Amidst the steady flow of research solutions stemming from the NLP domain, the study of use cases has fallen behind, leading to a number of innovative technical methods without a place in practice. In this work, we aim to build a structured overview of Legal Tech use cases, grounded in NLP literature, but also supplemented by voices from legal practice in Germany. Based upon a Systematic Literature Review, we identify seven categories of NLP technologies for the legal domain, which are then studied in juxtaposition to 22 legal use cases. In the investigation of these use cases, we identify 15 ethical, legal, and social aspects (ELSA), shedding light on the potential concerns of digitally transforming the legal domain.

cs.CL↗

arXiv2024

1-Diffractor: Efficient and Utility-Preserving Text Obfuscation Leveraging Word-Level Metric Differential Privacy

The study of privacy-preserving Natural Language Processing (NLP) has gained rising attention in recent years. One promising avenue studies the integration of Differential Privacy in NLP, which has brought about innovative methods in a variety of application settings. Of particular note are $\textit{word-level Metric Local Differential Privacy (MLDP)}$ mechanisms, which work to obfuscate potentially sensitive input text by performing word-by-word $\textit{perturbations}$. Although these methods have shown promising results in empirical tests, there are two major drawbacks: (1) the inevitable loss of utility due to addition of noise, and (2) the computational expensiveness of running these mechanisms on high-dimensional word embeddings. In this work, we aim to address these challenges by proposing $\texttt{1-Diffractor}$, a new mechanism that boasts high speedups in comparison to previous mechanisms, while still demonstrating strong utility- and privacy-preserving capabilities. We evaluate $\texttt{1-Diffractor}$ for utility on several NLP tasks, for theoretical and task-based privacy, and for efficiency in terms of speed and memory. $\texttt{1-Diffractor}$ shows significant improvements in efficiency, while still maintaining competitive utility and privacy scores across all conducted comparative tests against previous MLDP mechanisms. Our code is made available at: https://github.com/sjmeis/Diffractor.

cs.CL↗

arXiv2024

A Comparative Analysis of Word-Level Metric Differential Privacy: Benchmarking The Privacy-Utility Trade-off

The application of Differential Privacy to Natural Language Processing techniques has emerged in relevance in recent years, with an increasing number of studies published in established NLP outlets. In particular, the adaptation of Differential Privacy for use in NLP tasks has first focused on the $\textit{word-level}$, where calibrated noise is added to word embedding vectors to achieve "noisy" representations. To this end, several implementations have appeared in the literature, each presenting an alternative method of achieving word-level Differential Privacy. Although each of these includes its own evaluation, no comparative analysis has been performed to investigate the performance of such methods relative to each other. In this work, we conduct such an analysis, comparing seven different algorithms on two NLP tasks with varying hyperparameters, including the $\textit{epsilon ($\varepsilon$)}$ parameter, or privacy budget. In addition, we provide an in-depth analysis of the results with a focus on the privacy-utility trade-off, as well as open-source our implementation code for further reproduction. As a result of our analysis, we give insight into the benefits and challenges of word-level Differential Privacy, and accordingly, we suggest concrete steps forward for the research field.

cs.CL↗

arXiv2023

Identifying Practical Challenges in the Implementation of Technical Measures for Data Privacy Compliance

Modern privacy regulations provide a strict mandate for data processing entities to implement appropriate technical measures to demonstrate compliance. In practice, determining what measures are indeed "appropriate" is not trivial, particularly in light of vague guidelines provided by privacy regulations. To exacerbate the issue, challenges arise not only in the implementation of the technical measures themselves, but also in a variety of factors involving the roles, processes, decisions, and culture surrounding the pursuit of privacy compliance. In this paper, we present 33 challenges faced in the implementation of technical measures for privacy compliance, derived from a qualitative analysis of 16 interviews with privacy professionals. In addition, we evaluate the interview findings in a survey study, which gives way to a discussion of the identified challenges and their implications.

cs.CR↗

arXiv2023

Transforming Unstructured Text into Data with Context Rule Assisted Machine Learning (CRAML)

We describe a method and new no-code software tools enabling domain experts to build custom structured, labeled datasets from the unstructured text of documents and build niche machine learning text classification models traceable to expert-written rules. The Context Rule Assisted Machine Learning (CRAML) method allows accurate and reproducible labeling of massive volumes of unstructured text. CRAML enables domain experts to access uncommon constructs buried within a document corpus, and avoids limitations of current computational approaches that often lack context, transparency, and interpetability. In this research methods paper, we present three use cases for CRAML: we analyze recent management literature that draws from text data, describe and release new machine learning models from an analysis of proprietary job advertisement text, and present findings of social and economic interest from a public corpus of franchise documents. CRAML produces document-level coded tabular datasets that can be used for quantitative academic research, and allows qualitative researchers to scale niche classification schemes over massive text data. CRAML is a low-resource, flexible, and scalable methodology for building training data for supervised ML. We make available as open-source resources: the software, job advertisement text classifiers, a novel corpus of franchise documents, and a fully replicable start-to-finish trained example in the context of no poach clauses.

cs.CL↗

arXiv2022

Understanding the Implementation of Technical Measures in the Process of Data Privacy Compliance: A Qualitative Study

Modern privacy regulations, such as the General Data Protection Regulation (GDPR), address privacy in software systems in a technologically agnostic way by mentioning general "technical measures" for data privacy compliance rather than dictating how these should be implemented. An understanding of the concept of technical measures and how exactly these can be handled in practice, however, is not trivial due to its interdisciplinary nature and the necessary technical-legal interactions. We aim to investigate how the concept of technical measures for data privacy compliance is understood in practice as well as the technical-legal interaction intrinsic to the process of implementing those technical measures. We follow a research design that is 1) exploratory in nature, 2) qualitative, and 3) interview-based, with 16 selected privacy professionals in the technical and legal domains. Our results suggest that there is no clear mutual understanding and commonly accepted approach to handling technical measures. Both technical and legal roles are involved in the implementation of such measures. While they still often operate in separate spheres, a predominant opinion amongst the interviewees is to promote more interdisciplinary collaboration. Our empirical findings confirm the need for better interaction between legal and engineering teams when implementing technical measures for data privacy. We posit that interdisciplinary collaboration is paramount to a more complete understanding of technical measures, which currently lacks a mutually accepted notion. Yet, as strongly suggested by our results, there is still a lack of systematic approaches to such interaction. Therefore, the results strengthen our confidence in the need for further investigations into the technical-legal dynamic of data privacy compliance.

cs.SE↗

arXiv2022

Differential Privacy in Natural Language Processing: The Story So Far

As the tide of Big Data continues to influence the landscape of Natural Language Processing (NLP), the utilization of modern NLP methods has grounded itself in this data, in order to tackle a variety of text-based tasks. These methods without a doubt can include private or otherwise personally identifiable information. As such, the question of privacy in NLP has gained fervor in recent years, coinciding with the development of new Privacy-Enhancing Technologies (PETs). Among these PETs, Differential Privacy boasts several desirable qualities in the conversation surrounding data privacy. Naturally, the question becomes whether Differential Privacy is applicable in the largely unstructured realm of NLP. This topic has sparked novel research, which is unified in one basic goal: how can one adapt Differential Privacy to NLP methods? This paper aims to summarize the vulnerabilities addressed by Differential Privacy, the current thinking, and above all, the crucial next steps that must be considered.

cs.CL↗