Abstract Against the backdrop of the ever-improving Natural Language Inference (NLI) models, recent efforts have focused on the suitability of the current NLI datasets and on the feasibility of the NLI task as it is currently approached. Many of the recent studies have exposed the inherent human disagreements of the inference task and have proposed a shift from categorical labels to human subjective probability assessments, capturing human uncertainty. In this work, we show how neither the current task formulation nor the proposed uncertainty gradient are entirely suitable for solving the NLI challenges. Instead, we propose an ordered sense space annotation, which distinguishes between logical and common-sense inference. One end of the space captures non-sensical inferences, while the other end represents strictly logical scenarios. In the middle of the space, we find a continuum of common-sense, namely, the subjective and graded opinion of a “person on the street.” To arrive at the proposed annotation scheme, we perform a careful investigation of the SICK corpus and we create a taxonomy of annotation issues and guidelines. We re-annotate the corpus with the proposed annotation scheme, utilizing four symbolic inference systems, and then perform a thorough evaluation of the scheme by fine-tuning and testing commonly used pre-trained language models on the re-annotated SICK within various settings. We also pioneer a crowd annotation of a small portion of the MultiNLI corpus, showcasing that it is possible to adapt our scheme for annotation by non-experts on another NLI corpus. Our work shows the efficiency and benefits of the proposed mechanism and opens the way for a careful NLI task refinement.

With the success of contextualized language models, much research explores what these models really learn and in which cases they still fail. Most of this work focuses on specific NLP tasks and on the learning outcome. Little research has attempted to decouple the models’ weaknesses from specific tasks and focus on the embeddings per se and their mode of learning. In this paper, we take up this research opportunity: based on theoretical linguistic insights, we explore whether the semantic constraints of function words are learned and how the surrounding context impacts their embeddings. We create suitable datasets, provide new insights into the inner workings of LMs vis-a-vis function words and implement an assisting visual web interface for qualitative analysis.

Abstract Language models, such as BERT, construct multiple, contextualized embeddings for each word occurrence in a corpus. Understanding how the contextualization propagates through the model's layers is crucial for deciding which layers to use for a specific analysis task. Currently, most embedding spaces are explained by probing classifiers; however, some findings remain inconclusive. In this paper, we present LMFingerprints, a novel scoring‐based technique for the explanation of contextualized word embeddings. We introduce two categories of scoring functions, which measure (1) the degree of contextualization, i.e., the layerwise changes in the embedding vectors, and (2) the type of contextualization, i.e., the captured context information. We integrate these scores into an interactive explanation workspace. By combining visual and verbal elements, we provide an overview of contextualization in six popular transformer‐based language models. We evaluate hypotheses from the domain of computational linguistics, and our results not only confirm findings from related work but also reveal new aspects about the information captured in the embedding spaces. For instance, we show that while numbers are poorly contextualized, stopwords have an unexpected high contextualization in the models' upper layers, where their neighborhoods shift from similar functionality tokens to tokens that contribute to the meaning of the surrounding sentences.

Rita Sevastjanova, Aikaterini-Lida Kalouli, Christin Beck, Hanna Schäfer, Mennatallah El-Assady. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Despite the advances in Natural Language Inference through the training of massive deep models, recent work has revealed the generalization difficulties of such models, which fail to perform on adversarial datasets with challenging linguistic phenomena.Such phenomena, however, can be handled well by symbolic systems.Thus, we propose Hy-NLI, a hybrid system that learns to identify an NLI pair as linguistically challenging or not.Based on that, it uses its symbolic or deep learning component, respectively, to make the final inference decision.We show how linguistically less complex cases are best solved by robust state-of-the-art models, like BERT and XLNet, while hard linguistic phenomena are best handled by our implemented symbolic engine.Our thorough evaluation shows that our hybrid system achieves state-of-the-art performance across mainstream and adversarial datasets and opens the way for further research into the hybrid direction.