Reasoning and inference are central to human and artificial intelligence. Modeling inference in human language is very challenging. With the availability of large annotated data (Bowman et al., 2015), it has recently become feasible to train neural network based inference models, which have shown to be very effective. In this paper, we present a new state-of-the-art result, achieving the accuracy of 88.6% on the Stanford Natural Language Inference Dataset. Unlike the previous top models that use very complicated network architectures, we first demonstrate that carefully designing sequential inference models based on chain LSTMs can outperform all previous models. Based on this, we further show that by explicitly considering recursive architectures in both local inference modeling and inference composition, we achieve additional improvement. Particularly, incorporating syntactic parsing information contributes to our best result-it further improves the performance even when added to the already very strong model.

Here for the first time we present a shared task on detecting stance from tweets: given a tweet and a target entity (person, organization, etc.), automatic natural language systems must determine whether the tweeter is in favor of the given target, against the given target, or whether neither inference is likely. The target of interest may or may not be referred to in the tweet, and it may or may not be the target of opinion. Two tasks are proposed. Task A is a traditional supervised classification task where 70% of the annotated data for a target is used as training and the rest for testing. For Task B, we use as test data all of the instances for a new target (not used in task A) and no training data is provided. Our shared task received submissions from 19 teams for Task A and from 9 teams for Task B. The highest classification F-score obtained was 67.82 for Task A and 56.28 for Task B. However, systems found it markedly more difficult to infer stance towards the target of interest from tweets that express opinion towards another entity.

We describe a state-of-the-art sentiment analysis system that detects (a) the sentiment of short informal textual messages such as tweets and SMS (message-level task) and (b) the sentiment of a word or a phrase within a message (term-level task). The system is based on a supervised statistical text classification approach leveraging a variety of surface-form, semantic, and sentiment features. The sentiment features are primarily derived from novel high-coverage tweet-specific sentiment lexicons. These lexicons are automatically generated from tweets with sentiment-word hashtags and from tweets with emoticons. To adequately capture the sentiment of words in negated contexts, a separate sentiment lexicon is generated for negated words. The system ranked first in the SemEval-2013 shared task `Sentiment Analysis in Twitter' (Task 2), obtaining an F-score of 69.02 in the message-level task and 88.93 in the term-level task. Post-competition improvements boost the performance to an F-score of 70.45 (message-level task) and 89.50 (term-level task). The system also obtains state-of-the-art performance on two additional datasets: the SemEval-2013 SMS test set and a corpus of movie review excerpts. The ablation experiments demonstrate that the use of the automatically generated lexicons results in performance gains of up to 6.5 absolute percentage points.

Reviews depict sentiments of customers towards various aspects of a product or service. Some of these aspects can be grouped into coarser aspect categories. SemEval-2014 had a shared task (Task 4) on aspect-level sentiment analysis, with over 30 teams participated. In this paper, we describe our submissions, which stood first in detecting aspect categories, first in detecting sentiment towards aspect categories, third in detecting aspect terms, and first and second in detecting sentiment towards aspect terms in the laptop and restaurant domains, respectively.

The importance of a research article is routinely measured by counting how many times it has been cited. However, treating all citations with equal weight ignores the wide variety of functions that citations perform. We want to automatically identify the subset of references in a bibliography that have a central academic influence on the citing paper. For this purpose, we examine the effectiveness of a variety of features for determining the academic influence of a citation.By asking authors to identify the key references in their own work, we created a dataset in which citations were labeled according to their academic influence. Using automatic feature selection with supervised machine learning, we found a model for predicting academic influence that achieves good performance on this dataset using only four features.The best features, among those we evaluated, were features based on the number of times a reference is mentioned in the body of a citing paper. The performance of these features inspired us to design an influence-primed h-index (the hip-index). Unlike the conventional h-index, it weights citations by how many times a reference is mentioned. According to our experiments, the hip-index is a better indicator of researcher performance than the conventional h-index.