The most popular Cross-Document Event Coreference Resolution (CDEC) datasets fail to convey the true difficulty of the task, due to the lack of lexical diversity between coreferring event triggers (words or phrases that refer to an event). Furthermore, there is a dearth of event datasets for figurative language, limiting a crucial avenue of research in event comprehension. We address these two issues by introducing ECB+META, a lexically rich variant of Event Coref Bank Plus (ECB+) for CDEC on symbolic and metaphoric language. We use ChatGPT as a tool for the metaphoric transformation of sentences in the documents of ECB+, then tag the original event triggers in the transformed sentences in a semi-automated manner. In this way, we avoid the re-annotation of expensive coreference links. We present results that show existing methods that work well on ECB+ struggle with ECB+META, thereby paving the way for CDEC research on a much more challenging dataset. Code/data: https://github.com/ahmeshaf/llms_coref

Natural language generation tools are powerful and effective for generating content. However, language models are known to display bias and fairness issues, making them impractical to deploy for many use cases. We here focus on how fairness issues impact automatically generated test content, which can have stringent requirements to ensure the test measures only what it was intended to measure. Specifically, we review test content generated for a large-scale standardized English proficiency test with the goal of identifying content that only pertains to a certain subset of the test population as well as content that has the potential to be upsetting or distracting to some test takers. Issues like these could inadvertently impact a test taker’s score and thus should be avoided. This kind of content does not reflect the more commonly-acknowledged biases, making it challenging even for modern models that contain safeguards. We build a dataset of 601 generated texts annotated for fairness and explore a variety of methods for classification: fine-tuning, topic-based classification, and prompting, including few-shot and self-correcting prompts. We find that combining prompt self-correction and few-shot learning performs best, yielding an F1 score of 0.79 on our held-out test set, while much smaller BERT- and topic-based models have competitive performance on out-of-domain data.

Many Natural Language Processing (NLP) systems use annotated corpora for training and evaluation. However, labeled data is often costly to obtain and scaling annotation projects is difficult, which is why annotation tasks are often outsourced to paid crowdworkers. Citizen Science is an alternative to crowdsourcing that is relatively unexplored in the context of NLP. To investigate whether and how well Citizen Science can be applied in this setting, we conduct an exploratory study into engaging different groups of volunteers in Citizen Science for NLP by re-annotating parts of a pre-existing crowdsourced dataset. Our results show that this can yield high-quality annotations and at- tract motivated volunteers, but also requires considering factors such as scalability, participation over time, and legal and ethical issues. We summarize lessons learned in the form of guidelines and provide our code and data to aid future work on Citizen Science.

Natural language inference (NLI) has been widely used as a task to train and evaluate models for language understanding. However, the ability of NLI models to perform inferences requiring understanding of figurative language such as idioms and metaphors remains understudied. We introduce the IMPLI (Idiomatic and Metaphoric Paired Language Inference) dataset, an English dataset consisting of paired sentences spanning idioms and metaphors. We develop novel methods to generate 24k semiautomatic pairs as well as manually creating 1.8k gold pairs. We use IMPLI to evaluate NLI models based on RoBERTa fine-tuned on the widely used MNLI dataset. We then show that while they can reliably detect entailment relationship between figurative phrases with their literal counterparts, they perform poorly on similarly structured examples where pairs are designed to be non-entailing. This suggests the limits of current NLI models with regard to understanding figurative language and this dataset serves as a benchmark for future improvements in this direction.

This work aims to employ natural language generation (NLG) to rapidly generate items for English language learning applications: this requires both language models capable of generating fluent, high-quality English, and to control the output of the generation to match the requirements of the relevant items. We experiment with deep pretrained models for this task, developing novel methods for controlling items for factors relevant in language learning: diverse sentences for different proficiency levels and argument structure to test grammar. Human evaluation demonstrates high grammatically scores for all models (3.4 and above out of 4), and higher length (24%) and complexity (9%) over the baseline for the advanced proficiency model. Our results show that we can achieve strong performance while adding additional control to ensure diverse, tailored content for individual users.

Language use changes over time, and this impacts the effectiveness of NLP systems. This phenomenon is even more prevalent in social media data during crisis events where meaning and frequency of word usage may change over the course of days. Contextual language models fail to adapt temporally, emphasizing the need for temporal adaptation in models which need to be deployed over an extended period of time. While existing approaches consider data spanning large periods of time (from years to decades), shorter time spans are critical for crisis data. We quantify temporal degradation for this scenario and propose methods to cope with performance loss by leveraging techniques from domain adaptation. To the best of our knowledge, this is the first effort to explore effects of rapid language change driven by adversarial adaptations, particularly during natural and human-induced disasters. Through extensive experimentation on diverse crisis datasets, we analyze under what conditions our approaches outperform strong baselines while highlighting the current limitations of temporal adaptation methods in scenarios where access to unlabeled data is scarce.

Generating metaphors is a difficult task as it requires understanding nuanced relationships between abstract concepts. In this paper, we aim to generate a metaphoric sentence given a literal expression by replacing relevant verbs. Guided by conceptual metaphor theory, we propose to control the generation process by encoding conceptual mappings between cognitive domains to generate meaningful metaphoric expressions. To achieve this, we develop two methods: 1) using FrameNet-based embeddings to learn mappings between domains and applying them at the lexical level (CM-Lex), and 2) deriving source/target pairs to train a controlled seq-to-seq generation model (CM-BART). We assess our methods through automatic and human evaluation for basic metaphoricity and conceptual metaphor presence. We show that the unsupervised CM-Lex model is competitive with recent deep learning metaphor generation systems, and CM-BART outperforms all other models both in automatic and human evaluations.

Metaphor generation is a difficult task, and has seen tremendous improvement with the advent of deep pretrained models. We focus here on the specific task of metaphoric paraphrase generation, in which we provide a literal sentence and generate a metaphoric sentence which paraphrases that input. We compare naive, “free” generation models with those that exploit forms of control over the generation process, adding additional information based on conceptual metaphor theory. We evaluate two methods for generating paired training data, which is then used to train T5 models for free and controlled generation. We use crowdsourcing to evaluate the results, showing that free models tend to generate more fluent paraphrases, while controlled models are better at generating novel metaphors. We then analyze evaluation metrics, showing that different metrics are necessary to capture different aspects of metaphoric paraphrasing. We release our data and models, as well as our annotated results in order to facilitate development of better evaluation metrics.

The SemLink resource provides mappings between a variety of lexical semantic ontologies, each with their strengths and weaknesses. To take advantage of these differences, the ability to move between resources is essential. This work describes advances made to improve the usability of the SemLink resource: the automatic addition of new instances and mappings, manual corrections, sense-based vectors and collocation information, and architecture built to automatically update the resource when versions of the underlying resources change. These updates improve coverage, provide new tools to leverage the capabilities of these resources, and facilitate seamless updates, ensuring the consistency and applicability of these mappings in the future.

In the field of metaphor detection, deep learning systems are the ubiquitous and achieve strong performance on many tasks. However, due to the complicated procedures for manually identifying metaphors, the datasets available are relatively small and fraught with complications. We show that using syntactic features and lexical resources can automatically provide additional high-quality training data for metaphoric language, and this data can cover gaps and inconsistencies in metaphor annotation, improving state-of-the-art word-level metaphor identification. This novel application of automatically improving training data improves classification across numerous tasks, and reconfirms the necessity of high-quality data for deep learning frameworks.

Identification of metaphoric language in text is critical for generating effective semantic representations for natural language understanding. Computational approaches to metaphor identification have largely relied on heuristic based models or feature-based machine learning, using hand-crafted lexical resources coupled with basic syntactic information. However, recent work has shown the predictive power of syntactic constructions in determining metaphoric source and target domains (Sullivan 2013). Our work intends to explore syntactic constructions and their relation to metaphoric language. We undertake a corpus-based analysis of predicate-argument constructions and their metaphoric properties, and attempt to effectively represent syntactic constructions as features for metaphor processing, both in identifying source and target domains and in distinguishing metaphoric words from non-metaphoric.

A large amount of social media data is generated during natural disasters, and identifying the relevant portions of this data is critical for researchers attempting to understand human behavior, the effects of information sources, and preparatory actions undertaken during these events. In order to classify human behavior during hazard events, we employ machine learning for two tasks: identifying hurricane related tweets and classifying user evacuation behavior during hurricanes. We show that feature-based and deep learning methods provide different benefits for tweet classification, and ensemble-based methods using linguistic, temporal, and geospatial features can effectively classify user behavior.

When a hazard such as a hurricane threatens, people are forced to make a wide variety of decisions, and the information they receive and produce can influence their own and others’ actions. As social media grows more popular, an increasing number of people are using social media platforms to obtain and share information about approaching threats and discuss their interpretations of the threat and their protective decisions. This work aims to improve understanding of natural disasters through social media and provide an annotation scheme to identify themes in user’s social media behavior and facilitate efforts in supervised machine learning. To that end, this work has three contributions: (1) the creation of an annotation scheme to consistently identify hazard-related themes in Twitter, (2) an overview of agreement rates and difficulties in identifying annotation categories, and (3) a public release of both the dataset and guidelines developed from this scheme.