Tasks, Datasets and Evaluation Metrics are important concepts for understanding experimental scientific papers. However, previous work on information extraction for scientific literature mainly focuses on the abstracts only, and does not treat datasets as a separate type of entity (Zadeh and Schumann, 2016; Luan et al., 2018). In this paper, we present a new corpus that contains domain expert annotations for Task (T), Dataset (D), Metric (M) entities 2,000 sentences extracted from NLP papers. We report experiment results on TDM extraction using a simple data augmentation strategy and apply our tagger to around 30,000 NLP papers from the ACL Anthology. The corpus is made publicly available to the community for fostering research on scientific publication summarization (Erera et al., 2019) and knowledge discovery.

Due to the fast pace at which research reports in behaviour change are published, researchers, consultants and policymakers would benefit from more automatic ways to process these reports. Automatic extraction of the reports’ intervention content, population, settings and their results etc. are essential in synthesising and summarising the literature. However, to the best of our knowledge, no unique resource exists at the moment to facilitate this synthesis. In this paper, we describe the construction of a corpus of published behaviour change intervention evaluation reports aimed at smoking cessation. We also describe and release the annotation of 57 entities, that can be used as an off-the-shelf data resource for tasks such as entity recognition, etc. Both the corpus and the annotation dataset are being made available to the community.

While the fast-paced inception of novel tasks and new datasets helps foster active research in a community towards interesting directions, keeping track of the abundance of research activity in different areas on different datasets is likely to become increasingly difficult. The community could greatly benefit from an automatic system able to summarize scientific results, e.g., in the form of a leaderboard. In this paper we build two datasets and develop a framework (TDMS-IE) aimed at automatically extracting task, dataset, metric and score from NLP papers, towards the automatic construction of leaderboards. Experiments show that our model outperforms several baselines by a large margin. Our model is a first step towards automatic leaderboard construction, e.g., in the NLP domain.

We present a novel system providing summaries for Computer Science publications. Through a qualitative user study, we identified the most valuable scenarios for discovery, exploration and understanding of scientific documents. Based on these findings, we built a system that retrieves and summarizes scientific documents for a given information need, either in form of a free-text query or by choosing categorized values such as scientific tasks, datasets and more. Our system ingested 270,000 papers, and its summarization module aims to generate concise yet detailed summaries. We validated our approach with human experts.

Having an understanding of interpersonal relationships is helpful in many contexts. Our system seeks to assist humans with that task, using textual information (e.g., case notes, speech transcripts, posts, books) as input. Specifically, our system first extracts qualitative and quantitative information elements (which we call signals) about interactions among persons, aggregates those to provide a condensed view of relationships and then enables users to explore all facets of the resulting social (multi-)graph through a visual interface.

Sentiment composition is a fundamental sentiment analysis problem. Previous work relied on manual rules and manually-created lexical resources such as negator lists, or learned a composition function from sentiment-annotated phrases or sentences. We propose a new approach for learning sentiment composition from a large, unlabeled corpus, which only requires a word-level sentiment lexicon for supervision. We automatically generate large sentiment lexicons of bigrams and unigrams, from which we induce a set of lexicons for a variety of sentiment composition processes. The effectiveness of our approach is confirmed through manual annotation, as well as sentiment classification experiments with both phrase-level and sentence-level benchmarks.

This paper investigates how to improve performance on information extraction tasks by constraining and sequencing CRF-based approaches. We consider two different relation extraction tasks, both from the medical literature: dependence relations and probability statements. We explore whether adding constraints can lead to an improvement over standard CRF decoding. Results on our relation extraction tasks are promising, showing significant increases in performance from both (i) adding constraints to post-process the output of a baseline CRF, which captures “domain knowledge”, and (ii) further allowing flexibility in the application of those constraints by leveraging a binary classifier as a pre-processing step.

Stance classification is a core component in on-demand argument construction pipelines. Previous work on claim stance classification relied on background knowledge such as manually-composed sentiment lexicons. We show that both accuracy and coverage can be significantly improved through automatic expansion of the initial lexicon. We also developed a set of contextual features that further improves the state-of-the-art for this task.

In this paper, we address the problem of argument relation classification where argument units are from different texts. We design a joint inference method for the task by modeling argument relation classification and stance classification jointly. We show that our joint model improves the results over several strong baselines.

We investigate which distributional properties should be present in a tagset by examining different mappings of various current part-of-speech tagsets, looking at English, German, and Italian corpora. Given the importance of distributional information, we present a simple model for evaluating how a tagset mapping captures distribution, specifically by utilizing a notion of frames to capture the local context. In addition to an accuracy metric capturing the internal quality of a tagset, we introduce a way to evaluate the external quality of tagset mappings so that we can ensure that the mapping retains linguistically important information from the original tagset. Although most of the mappings we evaluate are motivated by linguistic concerns, we also explore an automatic, bottom-up way to define mappings, to illustrate that better distributional mappings are possible. Comparing our initial evaluations to POS tagging results, we find that more distributional tagsets can sometimes result in worse accuracy, underscring the need to carefully define the properties of a tagset.

Based on the idea that local contexts predict the same basic category across a language, we develop a simple method for comparing tagsets across corpora. The principle differences between tagsets are evidenced by variation in categories in one corpus in the same contexts where another corpus exhibits only a single tag. Such mismatches highlight differences in the definitions of tags which are crucial when porting technology from one annotation scheme to another.