Web tables contain a large amount of valuable knowledge and have inspired tabular language models aimed at tackling table interpretation (TI) tasks. In this paper, we analyse a widely used benchmark dataset for evaluation of TI tasks, particularly focusing on the entity linking task. Our analysis reveals that this dataset is overly simplified, potentially reducing its effectiveness for thorough evaluation and failing to accurately represent tables as they appear in the real-world. To overcome this drawback, we construct and annotate a new more challenging dataset. In addition to introducing the new dataset, we also introduce a novel problem aimed at addressing the entity linking task: named entity recognition within cells. Finally, we propose a prompting framework for evaluating the newly developed large language models (LLMs) on this novel TI task. We conduct experiments on prompting LLMs under various settings, where we use both random and similarity-based selection to choose the examples presented to the models. Our ablation study helps us gain insights into the impact of the few-shot examples. Additionally, we perform qualitative analysis to gain insights into the challenges encountered by the models and to understand the limitations of the proposed dataset.

The capabilities of large language models (LLMs) have been successfully applied in the context of table representation learning. The recently proposed tabular language models have reported state-of-the-art results across various tasks for table interpretation. However, a closer look into the datasets commonly used for evaluation reveals an entity leakage from the train set into the test set. Motivated by this observation, we explore adversarial attacks that represent a more realistic inference setup. Adversarial attacks on text have been shown to greatly affect the performance of LLMs, but currently, there are no attacks targeting tabular language models. In this paper, we propose an evasive entity-swap attack for the column type annotation (CTA) task. Our CTA attack is the first black-box attack on tables, where we employ a similarity-based sampling strategy to generate adversarial examples. The experimental results show that the proposed attack generates up to a 70% drop in performance.

A number of sub-symbolic approaches have been developed to provide insights into the GNN decision making process. These are first important steps on the way to explainability, but the generated explanations are often hard to understand for users that are not AI experts. To overcome this problem, we introduce a conceptual approach combining sub-symbolic and symbolic methods for human-centric explanations, that incorporate domain knowledge and causality. We furthermore introduce the notion of fidelity as a metric for evaluating how close the explanation is to the GNN's internal decision making process. The evaluation with a chemical dataset and ontology shows the explanatory value and reliability of our method.

Smart factories are equipped with machines that can sense their manufacturing environments, interact with each other, and control production processes. Smooth operation of such factories requires that the machines and engineering personnel that conduct their monitoring and diagnostics share a detailed common industrial knowledge about the factory, e.g., in the form of knowledge graphs. Creation and maintenance of such knowledge is expensive and requires automation. In this work we show how machine learning that is specifically tailored towards industrial applications can help in knowledge graph completion. In particular, we show how knowledge completion can benefit from event logs that are common in smart factories. We evaluate this on the knowledge graph from a real world-inspired smart factory with encouraging results.

The graph structure of biomedical data differs from those in typical knowledge graph benchmark tasks. A particular property of biomedical data is the presence of long-range dependencies, which can be captured by patterns described as logical rules. We propose a novel method that combines these rules with a neural multi-hop reasoning approach that uses reinforcement learning. We conduct an empirical study based on the real-world task of drug repurposing by formulating this task as a link prediction problem. We apply our method to the biomedical knowledge graph Hetionet and show that our approach outperforms several baseline methods.

Due to the vast amount of user-generated content in the emerging Web 2.0, there is a growing need for computational processing of sentiment analysis in documents. Most of the current research in this field is devoted to product reviews from websites. Microblogs and social networks pose even a greater challenge to sentiment classification. However, especially marketing and political campaigns leverage from opinions expressed on Twitter or other social communication platforms. The objects of interest in this paper are the presidential candidates of the Republican Party in the USA and their campaign topics. In this paper we introduce the combination of the noun phrases’ frequency and their PMI measure as constraint on aspect extraction. This compensates for sparse phrases receiving a higher score than those composed of high-frequency words. Evaluation shows that the meronymy relationship between politicians and their topics holds and improves accuracy of aspect extraction.