Natural language inference by deep learning method

Natural Language inference refers to the problem of determining the relationships between a premise and a hypothesis, it is an emerging area of natural language processing. The paper uses deep learning methods to complete natural language inference task. The dataset includes 3GPP dataset and SNLI dataset. Gensim library is used to get the word embeddings, there are 2 methods which are word2vec and doc2vec to map the sentence to array. 2 deep learning models DNNClassifier and Attention are implemented separately to classify the relationship between the proposals from the telecommunication area dataset. The highest accuracy of the experiment is 88% and we found that the quality of the dataset decided the upper bound of the accuracy.

Low-rank Gallus gallus domesticus chicks are better at transitive inference reasoning

A form of deductive reasoning, transitive inference, is thought to allow animals to infer relationships between members of a social group without having to remember all the interactions that occur. Such an ability means that animals can avoid direct confrontations which could be costly. Here we show that chicks perform a transitive inference task differently according to sex and rank. In female chicks, low-ranking birds performed better than did the highest ranked. Male chicks, however, showed an inverted U-shape of ability across rank, with the middle ranked chicks best able to perform the task. These results are explained according to the roles the sexes take within the group. This research directly links the abilities of transitive inference learning and social hierarchy formation and prompts further investigation into the role of both sex and rank within the dynamics of group living.

Morphological Awareness in L2 Chinese Reading Comprehension: Testing of Mediating Routes

This study aims to examine the contribution of morphological awareness to second language (L2) Chinese reading comprehension through potential mediating factors. Adult L2 Chinese learners (n = 447) participated in the study and completed two morphological awareness tasks (segmentation and discrimination), two vocabulary knowledge tasks (character knowledge and word-meaning knowledge), one lexical inference task, and one reading comprehension task. By testing alternative path models, this study identified the preferred model assuming the covariates of morphological awareness and vocabulary knowledge. Morphological awareness and vocabulary knowledge jointly contributed to L2 Chinese reading comprehension through lexical inference. The written modality of morphological awareness induced the activation of both morphological and orthographic information in print. The result suggests that morphological awareness (in the form of grapho-morphological knowledge) and vocabulary knowledge seem to be two parallel components under the same construct predicting Chinese reading comprehension. More importantly, this study underscores the intermediary effect of lexical inference in associating morphological awareness and reading comprehension in L2 Chinese learners.

Structure shapes the representation of a novel category

Concepts contain rich structures that support flexible semantic cognition. These structures can be characterized by patterns of feature covariation: certain clusters of features tend to occur in the same items (e.g., feathers, wings, can fly). Existing computational models demonstrate how this kind of structure can be leveraged to slowly learn the distinctions between categories, on developmental timescales. It is not clear whether and how we leverage feature structure to quickly learn a novel category. We thus investigated how the internal structure of a new category is extracted from experience and what kinds of representations guide this learning. We predicted that humans can leverage feature clusters within an individual category to benefit learning and that this relies on the rapid formation of distributed representations. Novel categories were designed with patterns of feature associations determined by carefully constructed graph structures (Modular, Random, and Lattice). In Experiment 1, a feature inference task using verbal stimuli revealed that Modular categories—containing clusters of reliably covarying features—were more easily learned than non-Modular categories. Experiment 2 replicated this effect using visual categories. In Experiment 3, a temporal statistical learning paradigm revealed that this Modular benefit persisted even when category structure was incidental to the task. We found that a neural network model employing distributed representations was able to account for the effects, whereas prototype and exemplar models could not. The findings constrain theories of category learning and of structure learning more broadly, suggesting that humans quickly form distributed representations that reflect coherent feature structure.

Ignorance implicatures of modified numerals

Modified numerals, such as at least three and more than five, are known to sometimes give rise to ignorance inferences. However, there is disagreement in the literature regarding the nature of these inferences, their context dependence, and differences between at least and more than. We present a series of experiments which sheds new light on these issues. Our results show that (a) the ignorance inferences of at least are more robust than those of more than, (b) the presence and strength of the ignorance inferences triggered by both at least and more than depends on the question under discussion (QUD), and (c) whether ignorance inferences are detected in a given experimental setting depends partly on the task that participants are asked to perform (e.g., an acceptability task versus an inference task). We offer an Optimality Theoretic account of these findings. In particular, the task effect is captured by assuming that in performing an acceptability task, participants take the speaker’s perspective in order to determine whether an expression is optimal given a certain epistemic state, while in performing an inference task they take the addressee’s perspective in order to determine what the most likely epistemic state of the speaker is given a certain expression. To execute the latter task in a fully rational manner, participants have to perform higher-order reasoning about alternative expressions the speaker could have used. Under the assumption that participants do not always perform such higher-order reasoning but also often resort to so-called unidirectional optimization, the task effect finds a natural explanation. This also allows us to relate our finding to asymmetries between comprehension and production that have been found in language acquisition.

Medial temporal and prefrontal cortices encode structural task representations at different levels of abstraction

Memory generalisations may be underpinned by either encoding- or retrieval-based mechanisms. We used a transitive inference task to investigate whether these generalisation mechanisms are influenced by progressive vs randomly interleaved training, and overnight consolidation. On consecutive days, participants learnt pairwise discriminations from two transitive hierarchies before being tested during fMRI. Inference performance was consistently better following progressive training, and for pairs further apart in the transitive hierarchy. BOLD pattern similarity correlated with hierarchical distances in the medial temporal lobe (MTL) and medial prefrontal cortex (MPFC). These results are consistent with the use of representations that directly encode structural relationships between different task features. Furthermore, BOLD patterns in MPFC were similar across the two independent hierarchies. We conclude that humans preferentially employ encoding-based mechanisms to store map-like relational codes that can be used for memory generalisation. While both MTL and MPFC support these representations, the MPFC encodes more abstract relational information.

TRIPDECODER: Study Travel Time Attributes and Route Preferences of Metro Systems from Smart Card Data

In this article, we target at recovering the exact routes taken by commuters inside a metro system that are not captured by an Automated Fare Collection (AFC) system and hence remain unknown. We strategically propose two inference tasks to handle the recovering, one to infer the travel time of each travel link that contributes to the total duration of any trip inside a metro network and the other to infer the route preferences based on historical trip records and the travel time of each travel link inferred in the previous inference task. As these two inference tasks have interrelationship, most of existing works perform these two tasks simultaneously. However, our solution TripDecoder adopts a totally different approach. TripDecoder fully utilizes the fact that there are some trips inside a metro system with only one practical route available. It strategically decouples these two inference tasks by only taking those trip records with only one practical route as the input for the first inference task of travel time and feeding the inferred travel time to the second inference task as an additional input, which not only improves the accuracy but also effectively reduces the complexity of both inference tasks. Two case studies have been performed based on the city-scale real trip records captured by the AFC systems in Singapore and Taipei to compare the accuracy and efficiency of TripDecoder and its competitors. As expected, TripDecoder has achieved the best accuracy in both datasets, and it also demonstrates its superior efficiency and scalability.

Comparative cognition in three understudied ungulate species: European bison, forest buffalos and giraffes

Background Comparative cognition has historically focused on a few taxa such as primates, birds or rodents. However, a broader perspective is essential to understand how different selective pressures affect cognition in different taxa, as more recently shown in several studies. Here we present the same battery of cognitive tasks to two understudied ungulate species with different socio-ecological characteristics, European bison (Bison bonasus) and forest buffalos (Syncerus caffer nanus), and we compare their performance to previous findings in giraffes (Giraffa camelopardalis). We presented subjects with an Object permanence task, Memory tasks with 30 and 60 s delays, two inference tasks based on acoustic cues (i.e. Acoustic inference tasks) and a control task to check for the use of olfactory cues (i.e. Olfactory task). Results Overall, giraffes outperformed bison and buffalos, and bison outperformed buffalos (that performed at chance level). All species performed better in the Object permanence task than in the Memory tasks and one of the Acoustic inference tasks (which they likely solved by relying on stimulus enhancement). Giraffes performed better than buffalos in the Shake full Acoustic inference task, but worse than bison and buffalos in the Shake empty Acoustic inference task. Conclusions In sum, our results are in line with the hypothesis that specific socio-ecological characteristics played a crucial role in the evolution of cognition, and that higher fission-fusion levels and larger dietary breadth are linked to higher cognitive skills. This study shows that ungulates may be an excellent model to test evolutionary hypotheses on the emergence of cognition.