Neural models for abstractive summarization tend to generate output that is fluent and well-formed but lacks semantic faithfulness, or factuality, with respect to the input documents. In this paper, we analyze the tradeoff between abstractiveness and factuality of generated summaries across multiple datasets and models, using extensive human evaluations of factuality. In our analysis, we visualize the rates of change in factuality as we gradually increase abstractiveness using a decoding constraint, and we observe that, while increased abstractiveness generally leads to a drop in factuality, the rate of factuality decay depends on factors such as the data that the system was trained on. We introduce two datasets with human factuality judgements; one containing 10.2k generated summaries with systematically varied degrees of abstractiveness; the other containing 4.2k summaries from five different summarization models. We propose new factuality metrics that adjust for the degree of abstractiveness, and we use them to compare the abstractiveness-adjusted factuality of previous summarization works, providing baselines for future work.
The web contains an abundance of user- generated content. While this content is useful for many applications, it poses many challenges due to the presence of offensive, biased, and overall toxic language. In this work, we present a system that identifies and classifies sexist content at different levels of granularity. Using transformer-based models, we explore the value of data augmentation, use of ensemble methods, and leverage in-context learning using foundation models to tackle the task. We evaluate the different components of our system both quantitatively and qualitatively. Our best systems achieve an F1 score of 0.84 for the binary classification task aiming to identify whether a given content is sexist or not and 0.64 and 0.47 for the two multi-class tasks that aim to identify the coarse and fine-grained types of sexism present in the given content respectively.
Recently, neural models have been leveraged to significantly improve the performance of information extraction from semi-structured websites. However, a barrier for continued progress is the small number of datasets large enough to train these models. In this work, we introduce the PLAtE (Pages of Lists Attribute Extraction) benchmark dataset as a challenging new web extraction task. PLAtE focuses on shopping data, specifically extractions from product review pages with multiple items encompassing the tasks of: (1) finding product list segmentation boundaries and (2) extracting attributes for each product. PLAtE is composed of 52,898 items collected from 6,694 pages and 156,014 attributes, making it the first large-scale list page web extraction dataset. We use a multi-stage approach to collect and annotate the dataset and adapt three state-of-the-art web extraction models to the two tasks comparing their strengths and weaknesses both quantitatively and qualitatively.
Motivated by suggested question generation in conversational news recommendation systems, we propose a model for generating question-answer pairs (QA pairs) with self-contained, summary-centric questions and length-constrained, article-summarizing answers. We begin by collecting a new dataset of news articles with questions as titles and pairing them with summaries of varying length. This dataset is used to learn a QA pair generation model producing summaries as answers that balance brevity with sufficiency jointly with their corresponding questions. We then reinforce the QA pair generation process with a differentiable reward function to mitigate exposure bias, a common problem in natural language generation. Both automatic metrics and human evaluation demonstrate these QA pairs successfully capture the central gists of the articles and achieve high answer accuracy.