The propagation of offensive content through social media channels has garnered attention of the research community. Multiple works have proposed various semantically related yet subtle distinct categories of offensive speech. In this work, we explore meta-earning approaches to leverage the diversity of offensive speech corpora to enhance their reliable and efficient detection. We propose a joint embedding architecture that incorporates the input's label and definition for classification via Prototypical Network. Our model achieves at least 75% of the maximal F1-score while using less than 10% of the available training data across 4 datasets. Our experimental findings also provide a case study of training strategies valuable to combat resource scarcity.

One of the challenges of natural language understanding is to deal with the subjectivity of sentences, which may express opinions and emotions that add layers of complexity and nuance. Sentiment analysis is a field that aims to extract and analyze these subjective elements from text, and it can be applied at different levels of granularity, such as document, paragraph, sentence, or aspect. Aspect-based sentiment analysis is a well-studied topic with many available data sets and models. However, there is no clear definition of what makes a sentence difficult for aspect-based sentiment analysis. In this paper, we explore this question by conducting an experiment with three data sets: "Laptops", "Restaurants", and "MTSC" (Multi-Target-dependent Sentiment Classification), and a merged version of these three datasets. We study the impact of domain diversity and syntactic diversity on difficulty. We use a combination of classifiers to identify the most difficult sentences and analyze their characteristics. We employ two ways of defining sentence difficulty. The first one is binary and labels a sentence as difficult if the classifiers fail to correctly predict the sentiment polarity. The second one is a six-level scale based on how many of the top five best-performing classifiers can correctly predict the sentiment polarity. We also define 9 linguistic features that, combined, aim at estimating the difficulty at sentence level.

This literature review gives an overview of current approaches to perform domain adaptation in a low-resource and approaches to perform multilingual semantic search in a low-resource setting. We developed a new typology to cluster domain adaptation approaches based on the part of dense textual information retrieval systems, which they adapt, focusing on how to combine them efficiently. We also explore the possibilities of combining multilingual semantic search with domain adaptation approaches for dense retrievers in a low-resource setting.

The increasing rate at which scientific knowledge is discovered and health claims shared online has highlighted the importance of developing efficient fact-checking systems for scientific claims. The usual setting for this task in the literature assumes that the documents containing the evidence for claims are already provided and annotated or contained in a limited corpus. This renders the systems unrealistic for real-world settings where knowledge sources with potentially millions of documents need to be queried to find relevant evidence. In this paper, we perform an array of experiments to test the performance of open-domain claim verification systems. We test the final verdict prediction of systems on four datasets of biomedical and health claims in different settings. While keeping the pipeline's evidence selection and verdict prediction parts constant, document retrieval is performed over three common knowledge sources (PubMed, Wikipedia, Google) and using two different information retrieval techniques. We show that PubMed works better with specialized biomedical claims, while Figure 1: The experimental setup of the study: scientific Wikipedia is more suited for everyday health claims are passed through a fixed pipeline using three concerns. Likewise, BM25 excels in retrieval different knowledge sources, resulting in different final precision, while semantic search in recall of relevant prediction performance (as measured by F1 score).

The discourse around conspiracy theories is currently thriving amidst the rampant misinformation in online environments. Research in this field has been focused on detecting conspiracy theories on social media, often relying on limited datasets. In this study, we present a novel methodology for constructing a Twitter dataset that encompasses accounts engaged in conspiracy-related activities throughout the year 2022. Our approach centers on data collection that is independent of specific conspiracy theories and information operations. Additionally, our dataset includes a control group comprising randomly selected users who can be fairly compared to the individuals involved in conspiracy activities. This comprehensive collection effort yielded a total of 15K accounts and 37M tweets extracted from their timelines. We conduct a comparative analysis of the two groups across three dimensions: topics, profiles, and behavioral characteristics. The results indicate that conspiracy and control users exhibit similarity in terms of their profile metadata characteristics. However, they diverge significantly in terms of behavior and activity, particularly regarding the discussed topics, the terminology used, and their stance on trending subjects. In addition, we find no significant disparity in the presence of bot users between the two groups. Finally, we develop a classifier to identify conspiracy users using features borrowed from bot, troll and linguistic literature. The results demonstrate a high accuracy level (with an F1 score of 0.94), enabling us to uncover the most discriminating features associated with conspiracy-related accounts.

The purpose of modeling document relevance for search engines is to rank better in subsequent searches. Document-specific historical click-through rates can be important features in a dynamic ranking system which updates as we accumulate more sample. This paper describes the properties of several such features, and tests them in controlled experiments. Extending the inverse propensity weighting method to documents creates an unbiased estimate of document relevance. This feature can approximate relevance accurately, leading to near-optimal ranking in ideal circumstances. However, it has high variance that is increasing with respect to the degree of position bias. Furthermore, inaccurate position bias estimation leads to poor performance. Under several scenarios this feature can perform worse than biased click-through rates. This paper underscores the need for accurate position bias estimation, and is unique in suggesting simultaneous use of biased and unbiased position bias features.

Information retrieval (IR) methods for KGQA consist of two stages: subgraph extraction and answer reasoning. We argue current subgraph extraction methods underestimate the importance of structural dependencies among evidence facts. We propose Evidence Pattern Retrieval (EPR) to explicitly model the structural dependencies during subgraph extraction. We implement EPR by indexing the atomic adjacency pattern of resource pairs. Given a question, we perform dense retrieval to obtain atomic patterns formed by resource pairs. We then enumerate their combinations to construct candidate evidence patterns. These evidence patterns are scored using a neural model, and the best one is selected to extract a subgraph for downstream answer reasoning. Experimental results demonstrate that the EPR-based approach has significantly improved the F1 scores of IR-KGQA methods by over 10 points on ComplexWebQuestions and achieves competitive performance on WebQuestionsSP.

Mainstream cross-lingual task-oriented dialogue (ToD) systems leverage the transfer learning paradigm by training a joint model for intent recognition and slot-filling in English and applying it, zero-shot, to other languages. We address a gap in prior research, which often overlooked the transfer to lower-resource colloquial varieties due to limited test data. Inspired by prior work on English varieties, we craft and manually evaluate perturbation rules that transform German sentences into colloquial forms and use them to synthesize test sets in four ToD datasets. Our perturbation rules cover 18 distinct language phenomena, enabling us to explore the impact of each perturbation on slot and intent performance. Using these new datasets, we conduct an experimental evaluation across six different transformers. Here, we demonstrate that when applied to colloquial varieties, ToD systems maintain their intent recognition performance, losing 6% (4.62 percentage points) in accuracy on average. However, they exhibit a significant drop in slot detection, with a decrease of 31% (21 percentage points) in slot F1 score. Our findings are further supported by a transfer experiment from Standard American English to synthetic Urban African American Vernacular English.

The automatic identification of offensive language such as hate speech is important to keep discussions civil in online communities. Identifying hate speech in multimodal content is a particularly challenging task because offensiveness can be manifested in either words or images or a juxtaposition of the two. This paper presents the MasonPerplexity submission for the Shared Task on Multimodal Hate Speech Event Detection at CASE 2024 at EACL 2024. The task is divided into two sub-tasks: sub-task A focuses on the identification of hate speech and sub-task B focuses on the identification of targets in text-embedded images during political events. We use an XLM-roBERTa-large model for sub-task A and an ensemble approach combining XLM-roBERTa-base, BERTweet-large, and BERT-base for sub-task B. Our approach obtained 0.8347 F1-score in sub-task A and 0.6741 F1-score in sub-task B ranking 3rd on both sub-tasks.

In this paper, we describe our participation in the mesinesp Task of the BioASQ biomedical semantic indexing challenge. The participating system follows an approach based solely on conventional information retrieval tools. We have evaluated various alternatives for extracting index terms from IBECS/LILACS documents in order to be stored in an Apache Lucene index. Those indexed representations are queried using the contents of the article to be annotated and a ranked list of candidate labels is created from the retrieved documents. We also have evaluated a sort of limited Label Powerset approach which creates meta-labels joining pairs of DeCS labels with high co-occurrence scores, and an alternative method based on label profile matching. Results obtained in official runs seem to confirm the suitability of this approach for languages like Spanish.