This paper describes the results of SemEval-2024 Task 5: Argument Reasoning in Civil Procedure, consisting of a single task on judging and reasoning about the answers to questions in U.S. civil procedure. The dataset for this task contains question, answer and explanation pairs taken from The Glannon Guide To Civil Procedure (Glannon, 2018). The task was to classify in a binary manner if the answer is a correct choice for the question or not. Twenty participants submitted their solutions, with the best results achieving a remarkable 82.31% F1-score. We summarize and analyze the results from all participating systems and provide an overview over the systems of 14 participants.

Recent developments in deep learning have led to great success in various natural language processing (NLP) tasks. However, these applications may involve data that contain sensitive information. Therefore, how to achieve good performance while also protecting the privacy of sensitive data is a crucial challenge in NLP. To preserve privacy, Differential Privacy (DP), which can prevent reconstruction attacks and protect against potential side knowledge, is becoming a de facto technique for private data analysis. In recent years, NLP in DP models (DP-NLP) has been studied from different perspectives, which deserves a comprehensive review. In this paper, we provide the first systematic review of recent advances in DP deep learning models in NLP. In particular, we first discuss some differences and additional challenges of DP-NLP compared with the standard DP deep learning. Then, we investigate some existing work on DP-NLP andpresent its recent developments from three aspects: gradient perturbation based methods, embedding vector perturbation based methods, and ensemble model based methods. We also discuss some challenges and future directions.

What is preventing us from building a NLP system that could help real people in real situations, for instance when they need legal advice but don’t understand law? This question is trickier than one might think, because legal systems vary from country to country, so do the law books, availability of data, and incomprehensibility of legalese. In this paper we focus Germany (which employs the civil-law system where, roughly speaking, interpretation of law codes dominates over precedence) and lay a foundational work to address the laymen’s legal question answering empirically. We create GerLayQA, a new dataset comprising of 21k laymen’s legal questions paired with answers from lawyers and grounded to concrete law book paragraphs. We experiment with a variety of retrieval and answer generation models and provide an in-depth analysis of limitations, which helps us to provide first empirical answers to the question above.

Neural machine translation (NMT) is a widely popular text generation task, yet there is a considerable research gap in the development of privacy-preserving NMT models, despite significant data privacy concerns for NMT systems. Differentially private stochastic gradient descent (DP-SGD) is a popular method for training machine learning models with concrete privacy guarantees; however, the implementation specifics of training a model with DP-SGD are not always clarified in existing models, with differing software libraries used and code bases not always being public, leading to reproducibility issues. To tackle this, we introduce DP-NMT, an open-source framework for carrying out research on privacy-preserving NMT with DP-SGD, bringing together numerous models, datasets, and evaluation metrics in one systematic software package. Our goal is to provide a platform for researchers to advance the development of privacy-preserving NMT systems, keeping the specific details of the DP-SGD algorithm transparent and intuitive to implement. We run a set of experiments on datasets from both general and privacy-related domains to demonstrate our framework in use. We make our framework publicly available and welcome feedback from the community.

Although the NLP community has adopted central differential privacy as a go-to framework for privacy-preserving model training or data sharing, the choice and interpretation of the key parameter, privacy budget 𝜀 that governs the strength of privacy protection, remains largely arbitrary. We argue that determining the 𝜀 value should not be solely in the hands of researchers or system developers, but must also take into account the actual people who share their potentially sensitive data. In other words: Would you share your instant messages for 𝜀 of 10? We address this research gap by designing, implementing, and conducting a behavioral experiment (311 lay participants) to study the behavior of people in uncertain decision-making situations with respect to privacy-threatening situations. Framing the risk perception in terms of two realistic NLP scenarios and using a vignette behavioral study help us determine what 𝜀 thresholds would lead lay people to be willing to share sensitive textual data – to our knowledge, the first study of its kind.

Most tasks in NLP require labeled data. Data labeling is often done on crowdsourcing platforms due to scalability reasons. However, publishing data on public platforms can only be done if no privacy-relevant information is included. Textual data often contains sensitive information like person names or locations. In this work, we investigate how removing personally identifiable information (PII) as well as applying differential privacy (DP) rewriting can enable text with privacy-relevant information to be used for crowdsourcing. We find that DP-rewriting before crowdsourcing can preserve privacy while still leading to good label quality for certain tasks and data. PII-removal led to good label quality in all examined tasks, however, there are no privacy guarantees given.

Protecting privacy in contemporary NLP models is gaining in importance. So does the need to mitigate social biases of such models. But can we have both at the same time? Existing research suggests that privacy preservation comes at the price of worsening biases in classification tasks. In this paper, we explore the extent to which this tradeoff really holds when we incorporate both privacy preservation and de-biasing techniques into training text generation models. How does improving the model along one dimension affect the other dimension as well as the utility of the model? We conduct an extensive set of experiments that include bias detection, privacy attacks, language modeling, and performance on downstream tasks.

Privatized text rewriting with local differential privacy (LDP) is a recent approach that enables sharing of sensitive textual documents while formally guaranteeing privacy protection to individuals. However, existing systems face several issues, such as formal mathematical flaws, unrealistic privacy guarantees, privatization of only individual words, as well as a lack of transparency and reproducibility. In this paper, we propose a new system ‘DP-BART’ that largely outperforms existing LDP systems. Our approach uses a novel clipping method, iterative pruning, and further training of internal representations which drastically reduces the amount of noise required for DP guarantees. We run experiments on five textual datasets of varying sizes, rewriting them at different privacy guarantees and evaluating the rewritten texts on downstream text classification tasks. Finally, we thoroughly discuss the privatized text rewriting approach and its limitations, including the problem of the strict text adjacency constraint in the LDP paradigm that leads to the high noise requirement.

This cutting-edge tutorial will help the NLP community to get familiar with current research in privacy-preserving methods. We will cover topics as diverse as membership inference, differential privacy, homomorphic encryption, or federated learning, all with typical applications to NLP. The goal is not only to draw the interest of the broader community, but also to present some typical use-cases and potential pitfalls in applying privacy-preserving methods to human language technologies.

Text rewriting with differential privacy (DP) provides concrete theoretical guarantees for protecting the privacy of individuals in textual documents. In practice, existing systems may lack the means to validate their privacy-preserving claims, leading to problems of transparency and reproducibility. We introduce DP-Rewrite, an open-source framework for differentially private text rewriting which aims to solve these problems by being modular, extensible, and highly customizable. Our system incorporates a variety of downstream datasets, models, pre-training procedures, and evaluation metrics to provide a flexible way to lead and validate private text rewriting research. To demonstrate our software in practice, we provide a set of experiments as a case study on the ADePT DP text rewriting system, detecting a privacy leak in its pre-training approach. Our system is publicly available, and we hope that it will help the community to make DP text rewriting research more accessible and transparent.

Preserving privacy in contemporary NLP models allows us to work with sensitive data, but unfortunately comes at a price. We know that stricter privacy guarantees in differentially-private stochastic gradient descent (DP-SGD) generally degrade model performance. However, previous research on the efficiency of DP-SGD in NLP is inconclusive or even counter-intuitive. In this short paper, we provide an extensive analysis of different privacy preserving strategies on seven downstream datasets in five different ‘typical’ NLP tasks with varying complexity using modern neural models based on BERT and XtremeDistil architectures. We show that unlike standard non-private approaches to solving NLP tasks, where bigger is usually better, privacy-preserving strategies do not exhibit a winning pattern, and each task and privacy regime requires a special treatment to achieve adequate performance.

Pre-training large transformer models with in-domain data improves domain adaptation and helps gain performance on the domain-specific downstream tasks. However, sharing models pre-trained on potentially sensitive data is prone to adversarial privacy attacks. In this paper, we asked to which extent we can guarantee privacy of pre-training data and, at the same time, achieve better downstream performance on legal tasks without the need of additional labeled data. We extensively experiment with scalable self-supervised learning of transformer models under the formal paradigm of differential privacy and show that under specific training configurations we can improve downstream performance without sacrifying privacy protection for the in-domain data. Our main contribution is utilizing differential privacy for large-scale pre-training of transformer language models in the legal NLP domain, which, to the best of our knowledge, has not been addressed before.

We present a new NLP task and dataset from the domain of the U.S. civil procedure. Each instance of the dataset consists of a general introduction to the case, a particular question, and a possible solution argument, accompanied by a detailed analysis of why the argument applies in that case. Since the dataset is based on a book aimed at law students, we believe that it represents a truly complex task for benchmarking modern legal language models. Our baseline evaluation shows that fine-tuning a legal transformer provides some advantage over random baseline models, but our analysis reveals that the actual ability to infer legal arguments remains a challenging open research question.

As privacy gains traction in the NLP community, researchers have started adopting various approaches to privacy-preserving methods. One of the favorite privacy frameworks, differential privacy (DP), is perhaps the most compelling thanks to its fundamental theoretical guarantees. Despite the apparent simplicity of the general concept of differential privacy, it seems non-trivial to get it right when applying it to NLP. In this short paper, we formally analyze several recent NLP papers proposing text representation learning using DPText (Beigi et al., 2019a,b; Alnasser et al., 2021; Beigi et al., 2021) and reveal their false claims of being differentially private. Furthermore, we also show a simple yet general empirical sanity check to determine whether a given implementation of a DP mechanism almost certainly violates the privacy loss guarantees. Our main goal is to raise awareness and help the community understand potential pitfalls of applying differential privacy to text representation learning.

Graph convolutional networks (GCNs) are a powerful architecture for representation learning on documents that naturally occur as graphs, e.g., citation or social networks. However, sensitive personal information, such as documents with people’s profiles or relationships as edges, are prone to privacy leaks, as the trained model might reveal the original input. Although differential privacy (DP) offers a well-founded privacy-preserving framework, GCNs pose theoretical and practical challenges due to their training specifics. We address these challenges by adapting differentially-private gradient-based training to GCNs and conduct experiments using two optimizers on five NLP datasets in two languages. We propose a simple yet efficient method based on random graph splits that not only improves the baseline privacy bounds by a factor of 2.7 while retaining competitive F1 scores, but also provides strong privacy guarantees of epsilon = 1.0. We show that, under certain modeling choices, privacy-preserving GCNs perform up to 90% of their non-private variants, while formally guaranteeing strong privacy measures.

Differential privacy provides a formal approach to privacy of individuals. Applications of differential privacy in various scenarios, such as protecting users’ original utterances, must satisfy certain mathematical properties. Our contribution is a formal analysis of ADePT, a differentially private auto-encoder for text rewriting (Krishna et al, 2021). ADePT achieves promising results on downstream tasks while providing tight privacy guarantees. Our proof reveals that ADePT is not differentially private, thus rendering the experimental results unsubstantiated. We also quantify the impact of the error in its private mechanism, showing that the true sensitivity is higher by at least factor 6 in an optimistic case of a very small encoder’s dimension and that the amount of utterances that are not privatized could easily reach 100% of the entire dataset. Our intention is neither to criticize the authors, nor the peer-reviewing process, but rather point out that if differential privacy applications in NLP rely on formal guarantees, these should be outlined in full and put under detailed scrutiny.

Evaluating the trustworthiness of a model’s prediction is essential for differentiating between ‘right for the right reasons’ and ‘right for the wrong reasons’. Identifying textual spans that determine the target label, known as faithful rationales, usually relies on pipeline approaches or reinforcement learning. However, such methods either require supervision and thus costly annotation of the rationales or employ non-differentiable models. We propose a differentiable training–framework to create models which output faithful rationales on a sentence level, by solely applying supervision on the target task. To achieve this, our model solves the task based on each rationale individually and learns to assign high scores to those which solved the task best. Our evaluation on three different datasets shows competitive results compared to a standard BERT blackbox while exceeding a pipeline counterpart’s performance in two cases. We further exploit the transparent decision–making process of these models to prefer selecting the correct rationales by applying direct supervision, thereby boosting the performance on the rationale–level.

Arguing without committing a fallacy is one of the main requirements of an ideal debate. But even when debating rules are strictly enforced and fallacious arguments punished, arguers often lapse into attacking the opponent by an ad hominem argument. As existing research lacks solid empirical investigation of the typology of ad hominem arguments as well as their potential causes, this paper fills this gap by (1) performing several large-scale annotation studies, (2) experimenting with various neural architectures and validating our working hypotheses, such as controversy or reasonableness, and (3) providing linguistic insights into triggers of ad hominem using explainable neural network architectures.

Reasoning is a crucial part of natural language argumentation. To comprehend an argument, one must analyze its warrant, which explains why its claim follows from its premises. As arguments are highly contextualized, warrants are usually presupposed and left implicit. Thus, the comprehension does not only require language understanding and logic skills, but also depends on common sense. In this paper we develop a methodology for reconstructing warrants systematically. We operationalize it in a scalable crowdsourcing process, resulting in a freely licensed dataset with warrants for 2k authentic arguments from news comments. On this basis, we present a new challenging task, the argument reasoning comprehension task. Given an argument with a claim and a premise, the goal is to choose the correct implicit warrant from two options. Both warrants are plausible and lexically close, but lead to contradicting claims. A solution to this task will define a substantial step towards automatic warrant reconstruction. However, experiments with several neural attention and language models reveal that current approaches do not suffice.

The classical view on argumentation, such that arguments are logical structures consisting of different distinguishable parts and that parties exchange arguments in a rational way, is prevalent in textbooks but nonexistent in the real world. Instead, argumentation is a multifaceted communication tool built upon humans’ capabilities to easily use common sense, emotions, and social context. As humans, we are pretty good at it. Computational Argumentation tries to tackle these phenomena but has a long and not so easy way to go. In this talk, I would like to shed a light on several recent attempts to deal with argumentation computationally, such as addressing argument quality, understanding argument reasoning, dealing with fallacies, and how should we never ever argue online.

A natural language argument is composed of a claim as well as reasons given as premises for the claim. The warrant explaining the reasoning is usually left implicit, as it is clear from the context and common sense. This makes a comprehension of arguments easy for humans but hard for machines. This paper summarizes the first shared task on argument reasoning comprehension. Given a premise and a claim along with some topic information, the goal was to automatically identify the correct warrant among two candidates that are plausible and lexically close, but in fact imply opposite claims. We describe the dataset with 1970 instances that we built for the task, and we outline the 21 computational approaches that participated, most of which used neural networks. The results reveal the complexity of the task, with many approaches hardly improving over the random accuracy of about 0.5. Still, the best observed accuracy (0.712) underlines the principle feasibility of identifying warrants. Our analysis indicates that an inclusion of external knowledge is key to reasoning comprehension.

Argument mining has become a popular research area in NLP. It typically includes the identification of argumentative components, e.g. claims, as the central component of an argument. We perform a qualitative analysis across six different datasets and show that these appear to conceptualize claims quite differently. To learn about the consequences of such different conceptualizations of claim for practical applications, we carried out extensive experiments using state-of-the-art feature-rich and deep learning systems, to identify claims in a cross-domain fashion. While the divergent conceptualization of claims in different datasets is indeed harmful to cross-domain classification, we show that there are shared properties on the lexical level as well as system configurations that can help to overcome these gaps.

An important skill in critical thinking and argumentation is the ability to spot and recognize fallacies. Fallacious arguments, omnipresent in argumentative discourse, can be deceptive, manipulative, or simply leading to ‘wrong moves’ in a discussion. Despite their importance, argumentation scholars and NLP researchers with focus on argumentation quality have not yet investigated fallacies empirically. The nonexistence of resources dealing with fallacious argumentation calls for scalable approaches to data acquisition and annotation, for which the serious games methodology offers an appealing, yet unexplored, alternative. We present Argotario, a serious game that deals with fallacies in everyday argumentation. Argotario is a multilingual, open-source, platform-independent application with strong educational aspects, accessible at www.argotario.net.

The goal of argumentation mining, an evolving research field in computational linguistics, is to design methods capable of analyzing people’s argumentation. In this article, we go beyond the state of the art in several ways. (i) We deal with actual Web data and take up the challenges given by the variety of registers, multiple domains, and unrestricted noisy user-generated Web discourse. (ii) We bridge the gap between normative argumentation theories and argumentation phenomena encountered in actual data by adapting an argumentation model tested in an extensive annotation study. (iii) We create a new gold standard corpus (90k tokens in 340 documents) and experiment with several machine learning methods to identify argument components. We offer the data, source codes, and annotation guidelines to the community under free licenses. Our findings show that argumentation mining in user-generated Web discourse is a feasible but challenging task.

Argumentation quality is viewed differently in argumentation theory and in practical assessment approaches. This paper studies to what extent the views match empirically. We find that most observations on quality phrased spontaneously are in fact adequately represented by theory. Even more, relative comparisons of arguments in practice correlate with absolute quality ratings based on theory. Our results clarify how the two views can learn from each other.

Large Web corpora containing full documents with permissive licenses are crucial for many NLP tasks. In this article we present the construction of 12 million-pages Web corpus (over 10 billion tokens) licensed under CreativeCommons license family in 50+ languages that has been extracted from CommonCrawl, the largest publicly available general Web crawl to date with about 2 billion crawled URLs. Our highly-scalable Hadoop-based framework is able to process the full CommonCrawl corpus on 2000+ CPU cluster on the Amazon Elastic Map/Reduce infrastructure. The processing pipeline includes license identification, state-of-the-art boilerplate removal, exact duplicate and near-duplicate document removal, and language detection. The construction of the corpus is highly configurable and fully reproducible, and we provide both the framework (DKPro C4CorpusTools) and the resulting data (C4Corpus) to the research community.

We present a new large dataset of 12403 context-sensitive verb relations manually annotated via crowdsourcing. These relations capture fine-grained semantic information between verb-centric propositions, such as temporal or entailment relations. We propose a novel semantic verb relation scheme and design a multi-step annotation approach for scaling-up the annotations using crowdsourcing. We employ several quality measures and report on agreement scores. The resulting dataset is available under a permissive CreativeCommons license at www.ukp.tu-darmstadt.de/data/verb-relations/. It represents a valuable resource for various applications, such as automatic information consolidation or automatic summarization.