Although existing machine reading comprehension models are making rapid progress on many datasets, they are far from robust. In this article, we propose an understanding-oriented machine reading comprehension model to address three kinds of robustness issues, which are over-sensitivity, over-stability, and generalization. Specifically, we first use a natural language inference module to help the model understand the accurate semantic meanings of input questions to address the issues of over-sensitivity and over-stability. Then, in the machine reading comprehension module, we propose a memory-guided multi-head attention method that can further well understand the semantic meanings of input questions and passages. Third, we propose a multi-language learning mechanism to address the issue of generalization. Finally, these modules are integrated with a multi-task learning-based method. We evaluate our model on three benchmark datasets that are designed to measure models’ robustness, including DuReader (robust) and two SQuAD-related datasets. Extensive experiments show that our model can well address the mentioned three kinds of robustness issues. And it achieves much better results than the compared state-of-the-art models on all these datasets under different evaluation metrics, even under some extreme and unfair evaluations. The source code of our work is available at https://github.com/neukg/RobustMRC .

Tagging based relational triple extraction methods are attracting growing research attention recently. However, most of these methods take a unidirectional extraction framework that first extracts all subjects and then extracts objects and relations simultaneously based on the subjects extracted. This framework has an obvious deficiency that it is too sensitive to the extraction results of subjects. To overcome this deficiency, we propose a bidirectional extraction framework based method that extracts triples based on the entity pairs extracted from two complementary directions. Concretely, we first extract all possible subject-object pairs from two paralleled directions. These two extraction directions are connected by a shared encoder component, thus the extraction features from one direction can flow to another direction and vice versa. By this way, the extractions of two directions can boost and complement each other. Next, we assign all possible relations for each entity pair by a biaffine model. During training, we observe that the share structure will lead to a convergence rate inconsistency issue which is harmful to performance. So we propose a share-aware learning mechanism to address it. We evaluate the proposed model on multiple benchmark datasets. Extensive experimental results show that the proposed model is very effective and it achieves state-of-the-art results on all of these datasets. Moreover, experiments show that both the proposed bidirectional extraction framework and the share-aware learning mechanism have good adaptability and can be used to improve the performance of other tagging based methods. The source code of our work is available at: https://github.com/neukg/BiRTE.

Most existing multi-document machine reading comprehension models mainly focus on understanding the interactions between the input question and documents, but ignore the following two kinds of understandings. First, to understand the semantic meaning of words in the input question and documents from the perspective of each other. Second, to understand the supporting cues for a correct answer from the perspective of intra-document and inter-documents. Ignoring these two kinds of important understandings would make the models overlook some important information that may be helpful for finding correct answers. To overcome this deficiency, we propose a deep understanding based model for multi-document machine reading comprehension. It has three cascaded deep understanding modules which are designed to understand the accurate semantic meaning of words, the interactions between the input question and documents, and the supporting cues for the correct answer. We evaluate our model on two large scale benchmark datasets, namely TriviaQA Web and DuReader. Extensive experiments show that our model achieves state-of-the-art results on both datasets.

Tagging based methods are one of the mainstream methods in relational triple extraction. However, most of them suffer from the class imbalance issue greatly. Here we propose a novel tagging based model that addresses this issue from following two aspects. First, at the model level, we propose a three-step extraction framework that can reduce the total number of samples greatly, which implicitly decreases the severity of the mentioned issue. Second, at the intra-model level, we propose a confidence threshold based cross entropy loss that can directly neglect some samples in the major classes. We evaluate the proposed model on NYT and WebNLG. Extensive experiments show that it can address the mentioned issue effectively and achieves state-of-the-art results on both datasets. The source code of our model is available at: https://github.com/neukg/ConCasRTE.

Many information sources are not just sequences of distinguishable symbols but rather have invariances governed by alternative counting paradigms such as permutations, combinations, and partitions. We consider an entire classification of these invariances called the twelvefold way in enumerative combinatorics and develop a method to characterize lossless compression limits. Explicit computations for all twelve settings are carried out for i.i.d. uniform and Bernoulli distributions. Comparisons among settings provide quantitative insight.

Neural conversation models have shown great potentials towards generating fluent and informative responses by introducing external background knowledge.Nevertheless, it is laborious to construct such knowledge-grounded dialogues, and existing models usually perform poorly when transfer to new domains with limited training samples.Therefore, building a knowledge-grounded dialogue system under the low-resource setting is a still crucial issue.In this paper, we propose a novel threestage learning framework based on weakly supervised learning which benefits from large scale ungrounded dialogues and unstructured knowledge base.To better cooperate with this framework, we devise a variant of Transformer with decoupled decoder which facilitates the disentangled learning of response generation and knowledge incorporation.Evaluation results on two benchmarks indicate that our approach can outperform other state-of-the-art methods with less training data, and even in zero-resource scenario, our approach still performs well.

Table filling based relational triple extraction methods are attracting growing research interests due to their promising performance and their abilities on extracting triples from complex sentences.However, this kind of methods are far from their full potential because most of them only focus on using local features but ignore the global associations of relations and of token pairs, which increases the possibility of overlooking some important information during triple extraction.To overcome this deficiency, we propose a global feature-oriented triple extraction model that makes full use of the mentioned two kinds of global associations.Specifically, we first generate a table feature for each relation.Then two kinds of global associations are mined from the generated table features.Next, the mined global associations are integrated into the table feature of each relation.This "generate-mine-integrate" process is performed multiple times so that the table feature of each relation is refined step by step.Finally, each relation's table is filled based on its refined table feature, and all triples linked to this relation are extracted based on its filled table.We evaluate the proposed model on three benchmark datasets.Experimental results show our model is effective and it achieves state-of-the-art results on all of these datasets.The source code of our work is available at: https://github.com/neukg/GRTE.

Knowledge graph embedding is an important task and it will benefit lots of downstream applications.Currently, deep neural networks based methods achieve state-of-the-art performance.However, most of these existing methods are very complex and need much time for training and inference.To address this issue, we propose a simple but effective atrous convolution based knowledge graph embedding method.Compared with existing state-of-the-art methods, our method has following main characteristics.First, it effectively increases feature interactions by using atrous convolutions.Second, to address the original information forgotten issue and vanishing/exploding gradient issue, it uses the residual learning method.Third, it has simpler structure but much higher parameter efficiency.We evaluate our method on six benchmark datasets with different evaluation metrics.Extensive experiments show that our model is very effective.On these diverse datasets, it achieves better results than the compared state-of-theart methods on most of evaluation metrics.The source codes of our model could be found at https://github.

Egocentric distances are often underestimated in virtual environments through head-mounted displays (HMDs). Previous studies suggest that peripheral vision can influence distance perception. Specifically, light in the periphery may improve distance judgments in HMDs. In this study, we conducted a series of experiments with varied peripheral treatments around the viewport. First, we found that the peripheral brightness significantly influences distance judgments when the periphery is brighter than a certain threshold, and found a possible range where the threshold was in. Second, we extended our previous research by changing the size of the peripheral treatment. A larger visual field (field of view of the HMD) resulted in significantly more accurate distance judgments compared to our original experiments with black peripheral treatment. Last, we found that applying a pixelated peripheral treatment can also improve distance judgments. The result implies that augmenting peripheral vision with secondary low-resolution displays may improve distance judgments in HMDs.

The rise of affordable head-mounted displays (HMDs) has raised questions about how to best design user interfaces for this technology. This paper focuses on the use of HMDs for home and office applications that require substantial text input. A physical keyboard is a familiar and effective text input device in normal desktop computing. But without additional camera technology, an HMD occludes all visual feedback about a user's hand position over the keyboard. We describe a system that assists HMD users in typing on a physical keyboard. Our system has a virtual keyboard assistant that provides visual feedback inside the HMD about a user's actions on the physical keyboard. It also provides powerful automatic correction of typing errors by extending a state-of-the-art touchscreen decoder. In a study with 24 participants, we found our virtual keyboard assistant enabled users to type more accurately on a visually-occluded keyboard. We found users wearing an HMD could type at over 40 words-per-minute while obtaining an error rate of less than 5%.

Numerous studies have reported underestimated egocentric distances in virtual environments through head-mounted displays (HMDs). However, it has been found that distance judgments made through Oculus Rift HMDs are much less compressed, and their relatively high device field of view (FOV) may play an important role. Some studies showed that applying constant white light in viewers' peripheral vision improved their distance judgments through HMDs. In this study, we examine the effects of the device FOV and the peripheral vision by performing a blind walking experiment through an Oculus Rift DK2 HMD with three different conditions. For the BlackFrame condition, we rendered a rectangular black frame to reduce the device field of view of the DK2 HMD to match an NVIS nVisor ST60 HMD. In the WhiteFrame and GreyFrame conditions, we changed the frame color to solid white and middle grey. From the results, we found that the distance judgments made through the black frame were significantly underestimated relative to the WhiteFrame condition. However, no significant differences were observed between the WhiteFrame and GreyFrame conditions. This result provides evidence that the device FOV and peripheral light could influence distance judgments in HMDs, and the degree of influence might not change proportionally with respect to the peripheral light brightness.

Distance perception is important for many virtual reality applications, and numerous studies have found underestimated egocentric distances in head-mounted display (HMD) based virtual environments. Applying minification to imagery displayed in HMDs is a method that can reduce or eliminate the underestimation [Kuhl et al. 2009; Zhang et al. 2012]. In a previous study, we measured distance judgments with direct blind walking through an Oculus Rift DK1 HMD and found that participants judged distance accurately in a calibrated condition, and minification caused subjects to overestimate distances [Li et al. 2014]. This article describes two experiments built on the previous study to examine distance judgments and minification with the Oculus Rift DK2 HMD (Experiment 1), and in the real world with a simulated HMD (Experiment 2). From the results, we found statistically significant distance underestimation with the DK2, but the judgments were more accurate than results typically reported in HMD studies. In addition, we discovered that participants made similar distance judgments with the DK2 and the simulated HMD. Finally, we found for the first time that minification had a similar impact on distance judgments in both virtual and real-world environments.

Distance perception is a crucial component for many virtual reality applications, and numerous studies have shown that egocentric distances are judged to be compressed in head-mounted display (HMD) systems. Geometric minification, a technique where the graphics are rendered with a field of view that larger than the HMD's field of view, is one known method of eliminating the distance compression [Kuhl et al. 2009; Zhang et al. 2012]. This study uses direct blind walking to determine how minification might impact distance judgments in the Oculus Rift HMD which has a significantly larger FOV than previous minification studies. Our results show that people were able to make accurate distance judgments in a calibrated condition and that geometric minification causes people to overestimate distances. Since this study shows that minification can impact wide FOV displays such as the Oculus, we discuss how it may be necessary to use calibration techniques which are more thorough than those described in this paper.