This work investigates the use of interactively updated label suggestions to improve upon the efficiency of gathering annotations on the task of opinion mining in German Covid-19 social media data. We develop guidelines to conduct a controlled annotation study with social science students and find that suggestions from a model trained on a small, expert-annotated dataset already lead to a substantial improvement - in terms of inter-annotator agreement(+.14 Fleiss' $\kappa$) and annotation quality - compared to students that do not receive any label suggestions. We further find that label suggestions from interactively trained models do not lead to an improvement over suggestions from a static model. Nonetheless, our analysis of suggestion bias shows that annotators remain capable of reflecting upon the suggested label in general. Finally, we confirm the quality of the annotated data in transfer learning experiments between different annotator groups. To facilitate further research in opinion mining on social media data, we release our collected data consisting of 200 expert and 2,785 student annotations.
Annotation studies often require annotators to familiarize themselves with the task, its annotation scheme, and the data domain. This can be overwhelming in the beginning, mentally taxing, and induce errors into the resulting annotations; especially in citizen science or crowd sourcing scenarios where domain expertise is not required and only annotation guidelines are provided. To alleviate these issues, we propose annotation curricula, a novel approach to implicitly train annotators. Our goal is to gradually introduce annotators into the task by ordering instances that are annotated according to a learning curriculum. To do so, we first formalize annotation curricula for sentence- and paragraph-level annotation tasks, define an ordering strategy, and identify well-performing heuristics and interactively trained models on three existing English datasets. We then conduct a user study with 40 voluntary participants who are asked to identify the most fitting misconception for English tweets about the Covid-19 pandemic. Our results show that using a simple heuristic to order instances can already significantly reduce the total annotation time while preserving a high annotation quality. Annotation curricula thus can provide a novel way to improve data collection. To facilitate future research, we further share our code and data consisting of 2,400 annotations.
Generating metaphors is a difficult task as it requires understanding nuanced relationships between abstract concepts. In this paper, we aim to generate a metaphoric sentence given a literal expression by replacing relevant verbs. Guided by conceptual metaphor theory, we propose to control the generation process by encoding conceptual mappings between cognitive domains to generate meaningful metaphoric expressions. To achieve this, we develop two methods: 1) using FrameNet-based embeddings to learn mappings between domains and applying them at the lexical level (CM-Lex), and 2) deriving source/target pairs to train a controlled seq-to-seq generation model (CM-BART). We assess our methods through automatic and human evaluation for basic metaphoricity and conceptual metaphor presence. We show that the unsupervised CM-Lex model is competitive with recent deep learning metaphor generation systems, and CM-BART outperforms all other models both in automatic and human evaluations.
Neural IR models have often been studied in homogeneous and narrow settings, which has considerably limited insights into their generalization capabilities. To address this, and to allow researchers to more broadly establish the effectiveness of their models, we introduce BEIR (Benchmarking IR), a heterogeneous benchmark for information retrieval. We leverage a careful selection of 17 datasets for evaluation spanning diverse retrieval tasks including open-domain datasets as well as narrow expert domains. We study the effectiveness of nine state-of-the-art retrieval models in a zero-shot evaluation setup on BEIR, finding that performing well consistently across all datasets is challenging. Our results show BM25 is a robust baseline and Reranking-based models overall achieve the best zero-shot performances, however, at high computational costs. In contrast, Dense-retrieval models are computationally more efficient but often underperform other approaches, highlighting the considerable room for improvement in their generalization capabilities. In this work, we extensively analyze different retrieval models and provide several suggestions that we believe may be useful for future work. BEIR datasets and code are available at https://github.com/UKPLab/beir.
Language usage changes over time, and this can impact the effectiveness of NLP systems. This work investigates methods for adapting to changing discourse during crisis events. We explore social media data during crisis, for which effective, time-sensitive methods are necessary. We experiment with two separate methods to accommodate changing data: temporal pretraining, which uses unlabeled data for the target time periods to train better language models, and a model of embedding shift based on tools for analyzing semantic change. This shift allows us to counteract temporal drift by normalizing incoming data based on observed patterns of language change. Simulating scenarios in which we lack access to incoming labeled data, we demonstrate the effectiveness of these methods for a wide variety of crises, showing we can improve performance by up to 8.0 F1 score for relevance classification across datasets.
In this paper, we introduce SciGen, a new challenge dataset for the task of reasoning-aware data-to-text generation consisting of tables from scientific articles and their corresponding descriptions. Describing scientific tables goes beyond the surface realization of the table content and requires reasoning over table values. The unique properties of SciGen are that (1) tables mostly contain numerical values, and (2) the corresponding descriptions require arithmetic reasoning. SciGen is therefore the first dataset that assesses the arithmetic reasoning capabilities of generation models on complex input structures, i.e., tables from scientific articles. We study the effectiveness of state-of-the-art data-to-text generation models on SciGen and evaluate the results using common metrics as well as human evaluation. Our results and analyses show that (a) while humans like to reason for describing scientific tables, the ability of state-of-the-art models is severely limited on this task, (b) while adding more training data improves the results, it is not the solution for reasoning-aware text generation, and (c) one of the main bottlenecks for this task is the lack of proper automatic evaluation metrics. The data, code, and annotations for human evaluation will be available at https://github.com/UKPLab/SciGen. SciGen opens new avenues for future research in reasoning-aware text generation and evaluation.
Intermediate task fine-tuning has been shown to culminate in large transfer gains across many NLP tasks. With an abundance of candidate datasets as well as pre-trained language models, it has become infeasible to run the cross-product of all combinations to find the best transfer setting. In this work we first establish that similar sequential fine-tuning gains can be achieved in adapter settings, and subsequently consolidate previously proposed methods that efficiently identify beneficial tasks for intermediate transfer learning. We experiment with a diverse set of 42 intermediate and 11 target English classification, multiple choice, question answering, and sequence tagging tasks. Our results show that efficient embedding based methods that rely solely on the respective datasets outperform computational expensive few-shot fine-tuning approaches. Our best methods achieve an average Regret@3 of less than 1% across all target tasks, demonstrating that we are able to efficiently identify the best datasets for intermediate training.
Question answering systems should help users to access knowledge on a broad range of topics and to answer a wide array of different questions. Most systems fall short of this expectation as they are only specialized in one particular setting, e.g., answering factual questions with Wikipedia data. To overcome this limitation, we propose composing multiple QA agents within a meta-QA system. We argue that there exist a wide range of specialized QA agents in literature. Thus, we address the central research question of how to effectively and efficiently identify suitable QA agents for any given question. We study both supervised and unsupervised approaches to address this challenge, showing that TWEAC - Transformer with Extendable Agent Classifiers - achieves the best performance overall with 94% accuracy. We provide extensive insights on the scalability of TWEAC, demonstrating that it scales robustly to over 100 QA agents with each providing just 1000 examples of questions they can answer.