Reasoning is key to many decision making processes. It requires consolidating a set of rule-like premises that are often associated with degrees of uncertainty and observations to draw conclusions. In this work, we address both the case where premises are specified as numeric probabilistic rules and situations in which humans state their estimates using words expressing degrees of certainty. Existing probabilistic reasoning datasets simplify the task, e.g., by requiring the model to only rank textual alternatives, by including only binary random variables, or by making use of a limited set of templates that result in less varied text.In this work, we present QUITE, a question answering dataset of real-world Bayesian reasoning scenarios with categorical random variables and complex relationships. QUITE provides high-quality natural language verbalizations of premises together with evidence statements and expects the answer to a question in the form of an estimated probability. We conduct an extensive set of experiments, finding that logic-based models outperform out-of-the-box large language models on all reasoning types (causal, evidential, and explaining-away). Our results provide evidence that neuro-symbolic models are a promising direction for improving complex reasoning. We release QUITE and code for training and experiments on Github.

Table Question Answering (TQA) aims at composing an answer to a question based on tabular data. While prior research has shown that TQA models lack robustness, understanding the underlying cause and nature of this issue remains predominantly unclear, posing a significant obstacle to the development of robust TQA systems. In this paper, we formalize three major desiderata for a fine-grained evaluation of robustness of TQA systems. They should (i) answer questions regardless of alterations in table structure, (ii) base their responses on the content of relevant cells rather than on biases, and (iii) demonstrate robust numerical reasoning capabilities. To investigate these aspects, we create and publish a novel TQA evaluation benchmark in English. Our extensive experimental analysis reveals that none of the examined state-of-the-art TQA systems consistently excels in these three aspects. Our benchmark is a crucial instrument for monitoring the behavior of TQA systems and paves the way for the development of robust TQA systems. We release our benchmark publicly.

Monitoring the threat landscape to be aware of actual or potential attacks is of utmost importance to cybersecurity professionals. Information about cyber threats is typically distributed using natural language reports. Natural language processing can help with managing this large amount of unstructured information, yet to date, the topic has received little attention. With this paper, we present AnnoCTR, a new CC-BY-SA-licensed dataset of cyber threat reports. The reports have been annotated by a domain expert with named entities, temporal expressions, and cybersecurity-specific concepts including implicitly mentioned techniques and tactics. Entities and concepts are linked to Wikipedia and the MITRE ATT&CK knowledge base, the most widely-used taxonomy for classifying types of attacks. Prior datasets linking to MITRE ATT&CK either provide a single label per document or annotate sentences out-of-context; our dataset annotates entire documents in a much finer-grained way. In an experimental study, we model the annotations of our dataset using state-of-the-art neural models. In our few-shot scenario, we find that for identifying the MITRE ATT&CK concepts that are mentioned explicitly or implicitly in a text, concept descriptions from MITRE ATT&CK are an effective source for training data augmentation.

Generative language models have recently shown remarkable success in generating answers to questions in a given textual context. However, these answers may suffer from hallucination, wrongly cite evidence, and spread misleading information. In this work, we address this problem by employing ChatGPT, a state-of-the-art generative model, as a machine-reading system. We ask it to retrieve answers to lexically varied and open-ended questions from trustworthy instructive texts. We introduce WHERE (WikiHow Evidence REtrieval), a new high-quality evaluation benchmark of a set of WikiHow articles exhaustively annotated with evidence sentences to questions that comes with a special challenge: All questions are about the article’s topic, but not all can be answered using the provided context. We interestingly find that when using a regular question-answering prompt, ChatGPT neglects to detect the unanswerable cases. When provided with a few examples, it learns to better judge whether a text provides answer evidence or not. Alongside this important finding, our dataset defines a new benchmark for evidence retrieval in question answering, which we argue is one of the necessary next steps for making large language models more trustworthy.

Many natural language processing (NLP) tasks are naturally imbalanced, as some target categories occur much more frequently than others in the real world. In such scenarios, current NLP models tend to perform poorly on less frequent classes. Addressing class imbalance in NLP is an active research topic, yet, finding a good approach for a particular task and imbalance scenario is difficult. In this survey, the first overview on class imbalance in deep-learning based NLP, we first discuss various types of controlled and real-world class imbalance. Our survey then covers approaches that have been explicitly proposed for class-imbalanced NLP tasks or, originating in the computer vision community, have been evaluated on them. We organize the methods by whether they are based on sampling, data augmentation, choice of loss function, staged learning, or model design. Finally, we discuss open problems and how to move forward.

Aspectual meaning refers to how the internal temporal structure of situations is presented. This includes whether a situation is described as a state or as an event, whether the situation is finished or ongoing, and whether it is viewed as a whole or with a focus on a particular phase. This survey gives an overview of computational approaches to modeling lexical and grammatical aspect along with intuitive explanations of the necessary linguistic concepts and terminology. In particular, we describe the concepts of stativity, telicity, habituality, perfective and imperfective, as well as influential inventories of eventuality and situation types. Aspect is a crucial component of semantics, especially for precise reporting of the temporal structure of situations, and future NLP approaches need to be able to handle and evaluate it systematically.

Scientific publications follow conventionalized rhetorical structures. Classifying the Argumentative Zone (AZ), e.g., identifying whether a sentence states a Motivation, a Result or Background information, has been proposed to improve processing of scholarly documents. In this work, we adapt and extend this idea to the domain of materials science research. We present and release a new dataset of 50 manually annotated research articles. The dataset spans seven sub-topics and is annotated with a materials-science focused multi-label annotation scheme for AZ. We detail corpus statistics and demonstrate high inter-annotator agreement. Our computational experiments show that using domain-specific pre-trained transformer-based text encoders is key to high classification performance. We also find that AZ categories from existing datasets in other domains are transferable to varying degrees.

Understanding causality is a core aspect of intelligence. The Event Causality Identification with Causal News Corpus Shared Task addresses two aspects of this challenge: Subtask 1 aims at detecting causal relationships in texts, and Subtask 2 requires identifying signal words and the spans that refer to the cause or effect, respectively. Our system, which is based on pre-trained transformers, stacked sequence tagging, and synthetic data augmentation, ranks third in Subtask 1 and wins Subtask 2 with an F1 score of 72.8, corresponding to a margin of 13 pp. to the second-best system.

Patent Landscaping, one of the central tasks of intellectual property management, includes selecting and grouping patents according to user-defined technical or application-oriented criteria. While recent transformer-based models have been shown to be effective for classifying patents into taxonomies such as CPC or IPC, there is yet little research on how to support real-world Patent Landscape Studies (PLSs) using natural language processing methods. With this paper, we release three labeled datasets for PLS-oriented classification tasks covering two diverse domains. We provide a qualitative analysis and report detailed corpus statistics.Most research on neural models for patents has been restricted to leveraging titles and abstracts. We compare strong neural and non-neural baselines, proposing a novel model that takes into account textual information from the patents’ full texts as well as embeddings created based on the patents’ CPC labels. We find that for PLS-oriented classification tasks, going beyond title and abstract is crucial, CPC labels are an effective source of information, and combining all features yields the best results.

Modal verbs (e.g., can, should or must) occur highly frequently in scientific articles. Decoding their function is not straightforward: they are often used for hedging, but they may also denote abilities and restrictions. Understanding their meaning is important for accurate information extraction from scientific text.To foster research on the usage of modals in this genre, we introduce the MIST (Modals In Scientific Text) dataset, which contains 3737 modal instances in five scientific domains annotated for their semantic, pragmatic, or rhetorical function. We systematically evaluate a set of competitive neural architectures on MIST. Transfer experiments reveal that leveraging non-scientific data is of limited benefit for modeling the distinctions in MIST. Our corpus analysis provides evidence that scientific communities differ in their usage of modal verbs, yet, classifiers trained on scientific data generalize to some extent to unseen scientific domains.

In this paper, we revisit the task of negation resolution, which includes the subtasks of cue detection (e.g. “not”, “never”) and scope resolution. In the context of previous shared tasks, a variety of evaluation metrics have been proposed. Subsequent works usually use different subsets of these, including variations and custom implementations, rendering meaningful comparisons between systems difficult. Examining the problem both from a linguistic perspective and from a downstream viewpoint, we here argue for a negation-instance based approach to evaluating negation resolution. Our proposed metrics correspond to expectations over per-instance scores and hence are intuitively interpretable. To render research comparable and to foster future work, we provide results for a set of current state-of-the-art systems for negation resolution on three English corpora, and make our implementation of the evaluation scripts publicly available.

The introduction of pre-trained transformer-based contextualized word embeddings has led to considerable improvements in the accuracy of graph-based parsers for frameworks such as Universal Dependencies (UD). However, previous works differ in various dimensions, including their choice of pre-trained language models and whether they use LSTM layers. With the aims of disentangling the effects of these choices and identifying a simple yet widely applicable architecture, we introduce STEPS, a new modular graph-based dependency parser. Using STEPS, we perform a series of analyses on the UD corpora of a diverse set of languages. We find that the choice of pre-trained embeddings has by far the greatest impact on parser performance and identify XLM-R as a robust choice across the languages in our study. Adding LSTM layers provides no benefits when using transformer-based embeddings. A multi-task training setup outputting additional UD features may contort results. Taking these insights together, we propose a simple but widely applicable parser architecture and configuration, achieving new state-of-the-art results (in terms of LAS) for 10 out of 12 diverse languages.

This paper presents our multilingual dependency parsing system as used in the IWPT 2021 Shared Task on Parsing into Enhanced Universal Dependencies. Our system consists of an unfactorized biaffine classifier that operates directly on fine-tuned XLM-R embeddings and generates enhanced UD graphs by predicting the best dependency label (or absence of a dependency) for each pair of tokens. To avoid sparsity issues resulting from lexicalized dependency labels, we replace lexical items in relations with placeholders at training and prediction time, later retrieving them from the parse via a hybrid rule-based/machine-learning system. In addition, we utilize model ensembling at prediction time. Our system achieves high parsing accuracy on the blind test data, ranking 3rd out of 9 with an average ELAS F1 score of 86.97.

In this paper, we address the representation of coordinate constructions in Enhanced Universal Dependencies (UD), where relevant dependency links are propagated from conjunction heads to other conjuncts. English treebanks for enhanced UD have been created from gold basic dependencies using a heuristic rule-based converter, which propagates only core arguments. With the aim of determining which set of links should be propagated from a semantic perspective, we create a large-scale dataset of manually edited syntax graphs. We identify several systematic errors in the original data, and propose to also propagate adjuncts. We observe high inter-annotator agreement for this semantic annotation task. Using our new manually verified dataset, we perform the first principled comparison of rule-based and (partially novel) machine-learning based methods for conjunction propagation for English. We show that learning propagation rules is more effective than hand-designing heuristic rules. When using automatic parses, our neural graph-parser based edge predictor outperforms the currently predominant pipelines using a basic-layer tree parser plus converters.

It is generally agreed upon in the natural language processing (NLP) community that ethics should be integrated into any curriculum. Being aware of and understanding the relevant core concepts is a prerequisite for following and participating in the discourse on ethical NLP. We here present ready-made teaching material in the form of slides and practical exercises on ethical issues in NLP, which is primarily intended to be integrated into introductory NLP or computational linguistics courses. By making this material freely available, we aim at lowering the threshold to adding ethics to the curriculum. We hope that increased awareness will enable students to identify potentially unethical behavior.

In this paper, we present a first attempt at enriching German Universal Dependencies (UD) treebanks with enhanced dependencies. Similarly to the converter for English (Schuster and Manning, 2016), we develop a rule-based system for deriving enhanced dependencies from the basic layer, covering three linguistic phenomena: relative clauses, coordination, and raising/control. For quality control, we manually correct or validate a set of 196 sentences, finding that around 90% of added relations are correct. Our data analysis reveals that difficulties arise mainly due to inconsistencies in the basic layer annotations. We show that the English system is in general applicable to German data, but that adapting to the particularities of the German treebanks and language increases precision and recall by up to 10%. Comparing the application of our converter on gold standard dependencies vs. automatic parses, we find that F1 drops by around 10% in the latter setting due to error propagation. Finally, an enhanced UD parser trained on a converted treebank performs poorly when evaluated against our annotations, indicating that more work remains to be done to create gold standard enhanced German treebanks.

This paper presents a new challenging information extraction task in the domain of materials science. We develop an annotation scheme for marking information on experiments related to solid oxide fuel cells in scientific publications, such as involved materials and measurement conditions. With this paper, we publish our annotation guidelines, as well as our SOFC-Exp corpus consisting of 45 open-access scholarly articles annotated by domain experts. A corpus and an inter-annotator agreement study demonstrate the complexity of the suggested named entity recognition and slot filling tasks as well as high annotation quality. We also present strong neural-network based models for a variety of tasks that can be addressed on the basis of our new data set. On all tasks, using BERT embeddings leads to large performance gains, but with increasing task complexity, adding a recurrent neural network on top seems beneficial. Our models will serve as competitive baselines in future work, and analysis of their performance highlights difficult cases when modeling the data and suggests promising research directions.

This paper adds to the ongoing discussion in the natural language processing community on how to choose a good development set. Motivated by the real-life necessity of applying machine learning models to different data distributions, we propose a clustering-based data splitting algorithm. It creates development (or test) sets which are lexically different from the training data while ensuring similar label distributions. Hence, we are able to create challenging cross-validation evaluation setups while abstracting away from performance differences resulting from label distribution shifts between training and test data. In addition, we present a Python-based tool for analyzing and visualizing data split characteristics and model performance. We illustrate the workings and results of our approach using a sentiment analysis and a patent classification task.

This paper presents our system at the IWPT 2020 Shared Task on Parsing into Enhanced Universal Dependencies. Using a biaffine classifier architecture (Dozat and Manning, 2017) which operates directly on finetuned RoBERTa embeddings, our parser generates enhanced UD graphs by predicting the best dependency label (or absence of a dependency) for each pair of tokens in the sentence. We address label sparsity issues by replacing lexical items in relations with placeholders at prediction time, later retrieving them from the parse in a rule-based fashion. In addition, we ensure structural graph constraints using a simple set of heuristics. On the English blind test data, our system achieves a very high parsing accuracy, ranking 1st out of 10 with an ELAS F1 score of 88.94%.

HDT-UD, the largest German UD treebank by a large margin, as well as the German-LIT treebank, currently do not analyze preposition-determiner contractions such as zum (= zu dem, “to the”) as multi-word tokens, which is inconsistent both with UD guidelines as well as other German UD corpora (GSD and PUD). In this paper, we show that harmonizing corpora with regard to this highly frequent phenomenon using a lookup-table based approach leads to a considerable increase in automatic parsing performance.

This paper addresses the automatic recognition of telicity, an aspectual notion. A telic event includes a natural endpoint (“she walked home”), while an atelic event does not (“she walked around”). Recognizing this difference is a prerequisite for temporal natural language understanding. In English, this classification task is difficult, as telicity is a covert linguistic category. In contrast, in Slavic languages, aspect is part of a verb’s meaning and even available in machine-readable dictionaries. Our contributions are as follows. We successfully leverage additional silver standard training data in the form of projected annotations from parallel English-Czech data as well as context information, improving automatic telicity classification for English significantly compared to previous work. We also create a new data set of English texts manually annotated with telicity.

We present LQVSumm, a corpus of about 2000 automatically created extractive multi-document summaries from the TAC 2011 shared task on Guided Summarization, which we annotated with several types of linguistic quality violations. Examples for such violations include pronouns that lack antecedents or ungrammatical clauses. We give details on the annotation scheme and show that inter-annotator agreement is good given the open-ended nature of the task. The annotated summaries have previously been scored for Readability on a numeric scale by human annotators in the context of the TAC challenge; we show that the number of instances of violations of linguistic quality of a summary correlates with these intuitively assigned numeric scores. On a system-level, the average number of violations marked in a system’s summaries achieves higher correlation with the Readability scores than current supervised state-of-the-art methods for assigning a single readability score to a summary. It is our hope that our corpus facilitates the development of methods that not only judge the linguistic quality of automatically generated summaries as a whole, but which also allow for detecting, labeling, and fixing particular violations in a text.

Suffix trees are data structures that can be used to index a corpus. In this paper, we explore how some properties of suffix trees naturally provide the functionality of an n-gram language model with variable n. We explain these properties of suffix trees, which we leverage for our Suffix Tree Language Model (STLM) implementation and explain how a suffix tree implicitly contains the data needed for n-gram language modeling. We also discuss the kinds of smoothing techniques appropriate to such a model. We then show that our suffix-tree language model implementation is competitive when compared to the state-of-the-art language model SRILM (Stolke, 2002) in statistical machine translation experiments.