Transformer-based models are now predominant in NLP.They outperform approaches based on static models in many respects.This success has in turn prompted research that reveals a number of biases in the language models generated by transformers.In this paper we utilize this research on biases to investigate to what extent transformer-based language models allow for extracting knowledge about object relations (X occurs in Y; X consists of Z; action A involves using X).To this end, we compare contextualized models with their static counterparts. We make this comparison dependent on the application of a number of similarity measures and classifiers.Our results are threefold:Firstly, we show that the models combined with the different similarity measures differ greatly in terms of the amount of knowledge they allow for extracting.Secondly, our results suggest that similarity measures perform much worse than classifier-based approaches.Thirdly, we show that, surprisingly, static models perform almost as well as contextualized models – in some cases even better.

We argue that mainly due to technical innovation in the landscape of annotation tools, a conceptual change in annotation models and processes is also on the horizon. It is diagnosed that these changes are bound up with multi-media and multi-perspective facilities of annotation tools, in particular when considering virtual reality (VR) and augmented reality (AR) applications, their potential ubiquitous use, and the exploitation of externally trained natural language pre-processing methods. Such developments potentially lead to a dynamic and exploratory heuristic construction of the annotation process. With TextAnnotator an annotation suite is introduced which focuses on multi-mediality and multi-perspectivity with an interoperable set of task-specific annotation modules (e.g., for word classification, rhetorical structures, dependency trees, semantic roles, and more) and their linkage to VR and mobile implementations. The basic architecture and usage of TextAnnotator is described and related to the above mentioned shifts in the field.

People’s visual perception is very pronounced and therefore it is usually no problem for them to describe the space around them in words. Conversely, people also have no problems imagining a concept of a described space. In recent years many efforts have been made to develop a linguistic concept for spatial and spatial-temporal relations. However, the systems have not really caught on so far, which in our opinion is due to the complex models on which they are based and the lack of available training data and automated taggers. In this paper we describe a project to support spatial annotation, which could facilitate annotation by its many functions, but also enrich it with many more information. This is to be achieved by an extension by means of a VR environment, with which spatial relations can be better visualized and connected with real objects. And we want to use the available data to develop a new state-of-the-art tagger and thus lay the foundation for future systems such as improved text understanding for Text2Scene.

Despite the great importance of the Latin language in the past, there are relatively few resources available today to develop modern NLP tools for this language. Therefore, the EvaLatin Shared Task for Lemmatization and Part-of-Speech (POS) tagging was published in the LT4HALA workshop. In our work, we dealt with the second EvaLatin task, that is, POS tagging. Since most of the available Latin word embeddings were trained on either few or inaccurate data, we trained several embeddings on better data in the first step. Based on these embeddings, we trained several state-of-the-art taggers and used them as input for an ensemble classifier called LSTMVoter. We were able to achieve the best results for both the cross-genre and the cross-time task (90.64% and 87.00%) without using additional annotated data (closed modality). In the meantime, we further improved the system and achieved even better results (96.91% on classical, 90.87% on cross-genre and 87.35% on cross-time).

Coreference resolution (CR) aims to find all spans of a text that refer to the same entity. The F1-Scores on these task have been greatly improved by new developed End2End-approaches and transformer networks. The inclusion of CR as a pre-processing step is expected to lead to improvements in downstream tasks. The paper examines this effect with respect to word embeddings. That is, we analyze the effects of CR on six different embedding methods and evaluate them in the context of seven lexical-semantic evaluation tasks and instantiation/hypernymy detection. Especially in the last tasks we hoped for a significant increase in performance. We show that all word embedding approaches do not benefit significantly from pronoun substitution. The measurable improvements are only marginal (around 0.5% in most test cases). We explain this result with the loss of contextual information, reduction of the relative occurrence of rare words and the lack of pronouns to be replaced.