Tag Correspondence Model for User Tag Suggestion

Some microblog services encourage users to annotate themselves with multiple tags, indicating their attributes and interests. User tags play an important role for personalized recommendation and information retrieval. In order to better understand the semantics of user tags, we propose Tag Correspondence Model （TCM） to identify complex correspondences of tags from the rich context of microblog users. The correspondence of a tag is referred to as a unique element in the context which is semantically correlated with this tag. In TCM, we divide the context of a microblog user into various sources （such as short messages, user profile, and neighbors）. With a collection of users with annotated tags, TCM can automatically learn the correspondences of user tags from multiple sources. With the learned correspondences, we are able to interpret implicit semantics of tags. Moreover, for the users who have not annotated any tags, TCM can suggest tags according to users＇ context information. Extensive experiments on a real-world dataset demonstrate that our method can efficiently identify correspondences of tags, which may eventually represent semantic meanings of tags.

Optimizing Non-Decomposable Evaluation Metrics for NeuralMachine Translation

While optimizing model parameters with respect to evaluation metrics has recently proven to benefit end to-end neural machine translation (NMT), the evaluation metrics used in the training are restricted to be defined at the sentence level to facilitate online learning algorithms. This is undesirable because the final evaluation metrics used in the testing phase are usually non-decomposable (i.e., they are defined at the corpus level and cannot be expressed as the sum of sentence-level metrics). To minimize the discrepancy between the training and the testing, we propose to extend the minimum risk training (MRT) algorithm to take non-decomposable corpus-level evaluation metrics into consideration while still keeping the advantages of online training. This can be done by calculating corpus-level evaluation metrics on a subset of training data at each step in online training. Experiments on Chinese-English and English-French translation show that our approach improves the correlation between training and testing and significantly outperforms the MRT algorithm using decomposable evaluation metrics.

Recent Advances on Neural Headline Generation

Recently, neural models have been proposed for headline generation by learning to map documents to headlines with recurrent neural network. In this work, we give a detailed introduction and comparison of existing work and recent improvements in neural headline generation, with particular attention on how encoders, decoders and neural model training strategies alter the overall performance of the headline generation system. Furthermore, we perform quantitative analysis of most existing neural headline generation systems and summarize several key factors that impact the performance of headline generation systems. Meanwhile, we carry on detailed error analysis to typical neural headline generation systems in order to gain more comprehension. Our results and conclusions are hoped to benefit future research studies.

Evidence in Support of Analogical Reasoning Improvements with Executive Attention Intervention in Healthy Young Adults

Analogical reasoning improvement is important in educational outcome improvement.Inspired by recent ideas and evidence,we applied anti-saccade task training as an executive attention intervention and tested whether it could improve analogical reasoning performance.A serial-task paradigm was applied where participants performed an anti-saccade followed by an analogical reasoning task including a perception condition.The experimental group finished the anti-saccade task in which the ratio of anti-saccade trials to pro-saccade trials was 5:1 while the counterpart was 1:1 in the active control group.Also,a blank control group was established where participants merely finished the analogical reasoning task.Event-related electroencephalographic(EEG)data were recorded when participants were performing the executive attention and analogical reasoning tasks.In addition,their resting state EEG was collected before and after the executive attention intervention.Behaviorally,the experimental group reacted significantly faster than the other two groups in analogical reasoning but not in perception.At the neural level,in the experimental group alone,the anti-saccade trials elicited a smaller N2 than pro-saccade trials and the resting alpha power was improved after executive attention intervention.No significant difference in P2 was found between the two groups in analogical reasoning or perception but the experimental group showed a larger late positive component than the active control group in analogical reasoning.We also found that the late positive component mediated the relationship between the N2 of anti-saccade trials and analogical reasoning reaction times in the experimental group.We further discussed the role of executive attention in the analogical reasoning process,which may pave the way for the future reliable improvement of fluid intelligence.

Neural Parse Combination

Analyzing the syntactic structure of natural languages by parsing is an important task in artificial intelligence. Due to the complexity of natural languages, individual parsers tend to make different yet complementary errors. We propose a neural network based approach to combine parses from different parsers to yield a more accurate parse than individual ones. Unlike conventional approaches, our method directly transforms linearized candidate parses into the ground-truth parse. Experiments on the Penn English Treebank show that the proposed method improves over a state-of-the-art parser combination approach significantly.