Transformers are arguably the main workhorse in recent natural language processing research. By definition, a Transformer is invariant with respect to reordering of the input. However, language is inherently sequential and word order is essential to the semantics and syntax of an utterance. In this article, we provide an overview and theoretical comparison of existing methods to incorporate position information into Transformer models. The objectives of this survey are to (1) showcase that position information in Transformer is a vibrant and extensive research area; (2) enable the reader to compare existing methods by providing a unified notation and systematization of different approaches along important model dimensions; (3) indicate what characteristics of an application should be taken into account when selecting a position encoding; and (4) provide stimuli for future research.

Combining a pretrained language model (PLM) with textual patterns has been shown to help in both zero- and few-shot settings. For zero-shot performance, it makes sense to design patterns that closely resemble the text seen during self-supervised pretraining because the model has never seen anything else. Supervised training allows for more flexibility. If we allow for tokens outside the PLM’s vocabulary, patterns can be adapted more flexibly to a PLM’s idiosyncrasies. Contrasting patterns where a “token” can be any continuous vector from those where a discrete choice between vocabulary elements has to be made, we call our method CONtinous pAtterNs (CONAN). We evaluate CONAN on two established benchmarks for lexical inference in context (LIiC) a.k.a. predicate entailment, a challenging natural language understanding task with relatively small training data. In a direct comparison with discrete patterns, CONAN consistently leads to improved performance, setting a new state of the art. Our experiments give valuable insights on the kind of pattern that enhances a PLM’s performance on LIiC and raise important questions regarding our understanding of PLMs using text patterns.

We present Graformer, a novel Transformer-based encoder-decoder architecture for graph-to-text generation. With our novel graph self-attention, the encoding of a node relies on all nodes in the input graph - not only direct neighbors - facilitating the detection of global patterns. We represent the relation between two nodes as the length of the shortest path between them. Graformer learns to weight these node-node relations differently for different attention heads, thus virtually learning differently connected views of the input graph. We evaluate Graformer on two popular graph-to-text generation benchmarks, AGENDA and WebNLG, where it achieves strong performance while using many fewer parameters than other approaches.

Lexical inference in context (LIiC) is the task of recognizing textual entailment between two very similar sentences, i.e., sentences that only differ in one expression. It can therefore be seen as a variant of the natural language inference task that is focused on lexical semantics. We formulate and evaluate the first approaches based on pretrained language models (LMs) for this task: (i) a few-shot NLI classifier, (ii) a relation induction approach based on handcrafted patterns expressing the semantics of lexical inference, and (iii) a variant of (ii) with patterns that were automatically extracted from a corpus. All our approaches outperform the previous state of the art, showing the potential of pretrained LMs for LIiC. In an extensive analysis, we investigate factors of success and failure of our three approaches.

Graph-to-text generation aims to generate fluent texts from graph-based data. In this paper, we investigate two recent pretrained language models (PLMs) and analyze the impact of different task-adaptive pretraining strategies for PLMs in graph-to-text generation. We present a study across three graph domains: meaning representations, Wikipedia knowledge graphs (KGs) and scientific KGs. We show that approaches based on PLMs BART and T5 achieve new state-of-the-art results and that task-adaptive pretraining strategies improve their performance even further. We report new state-of-the-art BLEU scores of 49.72 on AMR-LDC2017T10, 59.70 on WebNLG, and 25.66 on AGENDA datasets - a relative improvement of 31.8%, 4.5%, and 42.4%, respectively, with our models generating significantly more fluent texts than human references. In an extensive analysis, we identify possible reasons for the PLMs’ success on graph-to-text tasks. Our findings suggest that the PLMs benefit from similar facts seen during pretraining or fine-tuning, such that they perform well even when the input graph is reduced to a simple bag of node and edge labels.

Creating datasets manually by human annotators is a laborious task that can lead to biased and inhomogeneous labels. We propose a flexible, semi-automatic framework for labeling data for relation extraction. Furthermore, we provide a dataset of preprocessed sentences from the requirements engineering domain, including a set of automatically created as well as hand-crafted labels. In our case study, we compare the human and automatic labels and show that there is a substantial overlap between both annotations.

Self-Attention Networks (SANs) are an integral part of successful neural architectures such as Transformer (Vaswani et al., 2017), and thus of pretrained language models such as BERT (Devlin et al., 2019) or GPT-3 (Brown et al., 2020). Training SANs on a task or pretraining them on language modeling requires large amounts of data and compute resources. We are searching for modifications to SANs that enable faster learning, i.e., higher accuracies after fewer update steps. We investigate three modifications to SANs: direct position interactions, learnable temperature, and convoluted attention. When evaluating them on part-of-speech tagging, we find that direct position interactions are an alternative to position embeddings, and convoluted attention has the potential to speed up the learning process.

Knowledge graphs (KGs) can vary greatly from one domain to another. Therefore supervised approaches to both graph-to-text generation and text-to-graph knowledge extraction (semantic parsing) will always suffer from a shortage of domain-specific parallel graph-text data; at the same time, adapting a model trained on a different domain is often impossible due to little or no overlap in entities and relations. This situation calls for an approach that (1) does not need large amounts of annotated data and thus (2) does not need to rely on domain adaptation techniques to work well on different domains. To this end, we present the first approach to unsupervised text generation from KGs and show simultaneously how it can be used for unsupervised semantic parsing. We evaluate our approach on WebNLG v2.1 and a new benchmark leveraging scene graphs from Visual Genome. Our system outperforms strong baselines for both text<->graph conversion tasks without any manual adaptation from one dataset to the other. In additional experiments, we investigate the impact of using different unsupervised objectives.

We present SherLIiC, a testbed for lexical inference in context (LIiC), consisting of 3985 manually annotated inference rule candidates (InfCands), accompanied by (i) ~960k unlabeled InfCands, and (ii) ~190k typed textual relations between Freebase entities extracted from the large entity-linked corpus ClueWeb09. Each InfCand consists of one of these relations, expressed as a lemmatized dependency path, and two argument placeholders, each linked to one or more Freebase types. Due to our candidate selection process based on strong distributional evidence, SherLIiC is much harder than existing testbeds because distributional evidence is of little utility in the classification of InfCands. We also show that, due to its construction, many of SherLIiC’s correct InfCands are novel and missing from existing rule bases. We evaluate a large number of strong baselines on SherLIiC, ranging from semantic vector space models to state of the art neural models of natural language inference (NLI). We show that SherLIiC poses a tough challenge to existing NLI systems.

In this work, we propose a new model for aspect-based sentiment analysis. In contrast to previous approaches, we jointly model the detection of aspects and the classification of their polarity in an end-to-end trainable neural network. We conduct experiments with different neural architectures and word representations on the recent GermEval 2017 dataset. We were able to show considerable performance gains by using the joint modeling approach in all settings compared to pipeline approaches. The combination of a convolutional neural network and fasttext embeddings outperformed the best submission of the shared task in 2017, establishing a new state of the art.

We present a new method for estimating vector space representations of words: embedding learning by concept induction. We test this method on a highly parallel corpus and learn semantic representations of words in 1259 different languages in a single common space. An extensive experimental evaluation on crosslingual word similarity and sentiment analysis indicates that concept-based multilingual embedding learning performs better than previous approaches.