Semantic Textual Similarity (STS) measures the meaning similarity of sentences. Applications include machine translation (MT), summarization, generation, question answering (QA), short answer grading, semantic search, dialog and conversational systems. The STS shared task is a venue for assessing the current state-of-the-art. The 2017 task focuses on multilingual and cross-lingual pairs with one sub-track exploring MT quality estimation (MTQE) data. The task obtained strong participation from 31 teams, with 17 participating in all language tracks. We summarize performance and review a selection of well performing methods. Analysis highlights common errors, providing insight into the limitations of existing models. To support ongoing work on semantic representations, the STS Benchmark is introduced as a new shared training and evaluation set carefully selected from the corpus of English STS shared task data (2012-2017).

/pdf/semeval-2017-task-1-semantic-textual-similarity-multilingual-1su3vk2wkg.pdf

SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluation

/pdf/semeval-2017-task-1-semantic-textual-similarity-multilingual-51ojcf0apx.pdf

SemEval-2017 Task 1: Semantic Textual Similarity Multilingual and Crosslingual Focused Evaluation

Graph convolutional networks (GCNs) are a powerful deep learning approach for graph-structured data. Recently, GCNs and subsequent variants have shown superior performance in various application areas on real-world datasets. Despite their success, most of the current GCN models are shallow, due to the {\em over-smoothing} problem. In this paper, we study the problem of designing and analyzing deep graph convolutional networks. We propose the GCNII, an extension of the vanilla GCN model with two simple yet effective techniques: {\em Initial residual} and {\em Identity mapping}. We provide theoretical and empirical evidence that the two techniques effectively relieves the problem of over-smoothing. Our experiments show that the deep GCNII model outperforms the state-of-the-art methods on various semi- and full-supervised tasks. Code is available at this https URL .

/pdf/simple-and-deep-graph-convolutional-networks-587ny8z144.pdf

Simple and Deep Graph Convolutional Networks

Automatic elicitation of semantic information from natural language texts is an important research problem with many practical application areas. Especially after the recent proliferation of online content through channels such as social media sites, news portals, and forums; solutions to problems such as sentiment analysis, sarcasm/controversy/veracity/rumour/fake news detection, and argument mining gained increasing impact and significance, revealed with large volumes of related scientific publications. In this article, we tackle an important problem from the same family and present a survey of stance detection in social media posts and (online) regular texts. Although stance detection is defined in different ways in different application settings, the most common definition is “automatic classification of the stance of the producer of a piece of text, towards a target, into one of these three classes: {Favor, Against, Neither}.” Our survey includes definitions of related problems and concepts, classifications of the proposed approaches so far, descriptions of the relevant datasets and tools, and related outstanding issues. Stance detection is a recent natural language processing topic with diverse application areas, and our survey article on this newly emerging topic will act as a significant resource for interested researchers and practitioners.

Stance Detection: A Survey

Stance detection on social media is an emerging opinion mining paradigm for various social and political applications in which sentiment analysis may be sub-optimal. There has been a growing research interest for developing effective methods for stance detection methods varying among multiple communities including natural language processing, web science, and social computing, where each modeled stance detection in different ways. In this paper, we survey the work on stance detection across those communities and present an exhaustive review of stance detection techniques on social media, including the task definition, different types of targets in stance detection, features set used, and various machine learning approaches applied. Our survey reports state-of-the-art results on the existing benchmark datasets on stance detection, and discusses the most effective approaches. In addition, we explore the emerging trends and different applications of stance detection on social media, including opinion mining and prediction and recently using it for fake news detection. The study concludes by discussing the gaps in the current existing research and highlights the possible future directions for stance detection on social media.

Stance detection on social media: State of the art and trends

Identifying the political perspective shaping the way news events are discussed in the media is an important and challenging task. In this paper, we highlight the importance of contextualizing social information, capturing how this information is disseminated in social networks. We use Graph Convolutional Networks, a recently proposed neural architecture for representing relational information, to capture the documents’ social context. We show that social information can be used effectively as a source of distant supervision, and when direct supervision is available, even little social information can significantly improve performance.

/pdf/encoding-social-information-with-graph-convolutional-3sr32vmn3n.pdf

Encoding Social Information with Graph Convolutional Networks forPolitical Perspective Detection in News Media

Online debates can help provide valuable information about various perspectives on a wide range of issues. However, understanding the stances expressed in these debates is a highly challenging task, which requires modeling both textual content and users’ conversational interactions. Current approaches take a collective classification approach, which ignores the relationships between different debate topics. In this work, we suggest to view this task as a representation learning problem, and embed the text and authors jointly based on their interactions. We evaluate our model over the Internet Argumentation Corpus, and compare different approaches for structural information embedding. Experimental results show that our model can achieve significantly better results compared to previous competitive models.

Structured Representation Learning for Online Debate Stance Prediction

The way information is generated and disseminated has changed dramatically over the last decade. Identifying the political perspective shaping the way events are discussed in the media becomes more important due to the sharp increase in the number of news outlets and articles. Previous approaches usually only leverage linguistic information. However, news articles attempt to maintain credibility and seem impartial. Therefore, bias is introduced in subtle ways, usually by emphasizing different aspects of the story. In this paper, we propose a novel framework that considers entities mentioned in news articles and external knowledge about them, capturing the bias with respect to those entities. We explore different ways to inject entity information into the text model. Experiments show that our proposed framework achieves significant improvements over the standard text models, and is capable of identifying the difference in news narratives with different perspectives.

/pdf/mean-multi-head-entity-aware-attention-networkfor-political-3xrqiimoo5.pdf

MEAN: Multi-head Entity Aware Attention Networkfor Political Perspective Detection in News Media

We introduce DRAIL, a new declarative framework for specifying Deep Relational Models. Our framework separates structural considerations, which express domain knowledge, from the learning architecture to simplify the process of building complex structural models. We show the DRAIL formulation of two NLP tasks, Twitter Part-of-Speech tagging and Entity-Relation extraction. We compare the performance of different deep learning architectures for these structural learning tasks.

/pdf/introducing-drail-a-step-towards-declarative-deep-relational-1vqmi6m3su.pdf

Introducing DRAIL – a Step Towards Declarative Deep Relational Learning

This paper describes our proposed solution for SemEval 2017 Task 1: Semantic Textual Similarity (Daniel Cer and Specia, 2017). The task aims at measuring the degree of equivalence between sentences given in English. Performance is evaluated by computing Pearson Correlation scores between the predicted scores and human judgements. Our proposed system consists of two subsystems and one regression model for predicting STS scores. The two subsystems are designed to learn Paraphrase and Event Embeddings that can take the consideration of paraphrasing characteristics and sentence structures into our system. The regression model associates these embeddings to make the final predictions. The experimental result shows that our system acquires 0.8 of Pearson Correlation Scores in this task.

/pdf/purduenlp-at-semeval-2017-task-1-predicting-semantic-textual-605684d4q6.pdf

Chang Li

Papers

Encoding Social Information with Graph Convolutional Networks forPolitical Perspective Detection in News Media

Structured Representation Learning for Online Debate Stance Prediction

MEAN: Multi-head Entity Aware Attention Networkfor Political Perspective Detection in News Media

Introducing DRAIL – a Step Towards Declarative Deep Relational Learning

PurdueNLP at SemEval-2017 Task 1: Predicting Semantic Textual Similarity with Paraphrase and Event Embeddings