Specialized number representations in NLP have shown improvements on numerical reasoning tasks like arithmetic word problems and masked number prediction. But humans also use numeracy to make better sense of world concepts, e.g., you can seat 5 peopl
e in your room' but not 500. Does a better grasp of numbers improve a model's understanding of other concepts and words? This paper studies the effect of using six different number encoders on the task of masked word prediction (MWP), as a proxy for evaluating literacy. To support this investigation, we develop Wiki-Convert, a 900,000 sentence dataset annotated with numbers and units, to avoid conflating nominal and ordinal number occurrences. We find a significant improvement in MWP for sentences containing numbers, that exponent embeddings are the best number encoders, yielding over 2 points jump in prediction accuracy over a BERT baseline, and that these enhanced literacy skills also generalize to contexts without annotated numbers. We release all code at https://git.io/JuZXn.
In this work, we present our approach and findings for SemEval-2021 Task 5 - Toxic Spans Detection. The task's main aim was to identify spans to which a given text's toxicity could be attributed. The task is challenging mainly due to two constraints:
the small training dataset and imbalanced class distribution. Our paper investigates two techniques, semi-supervised learning and learning with Self-Adjusting Dice Loss, for tackling these challenges. Our submitted system (ranked ninth on the leader board) consisted of an ensemble of various pre-trained Transformer Language Models trained using either of the above-proposed techniques.
This paper describes our system for SemEval-2021 Task 4: Reading Comprehension of Abstract Meaning. To accomplish this task, we utilize the Knowledge-Enhanced Graph Attention Network (KEGAT) architecture with a novel semantic space transformation str
ategy. It leverages heterogeneous knowledge to learn adequate evidences, and seeks for an effective semantic space of abstract concepts to better improve the ability of a machine in understanding the abstract meaning of natural language. Experimental results show that our system achieves strong performance on this task in terms of both imperceptibility and nonspecificity.
Potentially idiomatic expressions (PIEs) are ambiguous between non-compositional idiomatic interpretations and transparent literal interpretations. For example, hit the road'' can have an idiomatic meaning corresponding to start a journey' or have a
literal interpretation. In this paper we propose a supervised model based on contextualized embeddings for predicting whether usages of PIEs are idiomatic or literal. We consider monolingual experiments for English and Russian, and show that the proposed model outperforms previous approaches, including in the case that the model is tested on instances of PIE types that were not observed during training. We then consider cross-lingual experiments in which the model is trained on PIE instances in one language, English or Russian, and tested on the other language. We find that the model outperforms baselines in this setting. These findings suggest that contextualized embeddings are able to learn representations that encode knowledge of idiomaticity that is not restricted to specific expressions, nor to a specific language.
UDify is the state-of-the-art language-agnostic dependency parser which is trained on a polyglot corpus of 75 languages. This multilingual modeling enables the model to generalize over unknown/lesser-known languages, thus leading to improved performa
nce on low-resource languages. In this work we used linguistic typology knowledge available in URIEL database, to improve the cross-lingual transferring ability of UDify even further.
Wireless Sensor Networks (WSNs) are deployed in adversarial environments and
used for critical applications such as battle field surveillance and medical monitoring, then
security weaknesses become a big concern. The severe resource constraints of
WSNs give
rise to the need for resource bound security solutions.
The Implicit Geographic Forwarding Protocol (IGF) is considered stateless, which
means that it does not contain any routing tables and does not depend on the knowledge of
the network topology, or on the presence or absence of the node in WSN. This protocol is
developed to provide a range of mechanisms that increase security in IGF. Thus it keeps
the dynamic connectivity features and provides effective defenses against potential attacks.
These mechanisms supported the security against several attacks as Black hole, Sybil and
Retransmission attacks, but the problem was the inability of mechanisms to deal with
physical attack.
This research deals with a detailed study of the SIGF-2 protocol and proposes an
improvement for it, in which we use the concept of deployment knowledge from random
key pool algorithm of keys management to defend against physical attack . The evaluation
of simulation results, with different parameters, proved that our proposal had improved the
studied protocol.
This research deals with values through studying opinions of philosophers who consider that values are absolute ,and compare it with opinions of philosophers who consider that values are relative. Then, the research analyzes and criticizes opinions a
nd arguments of the two parts to discover their knowledge ,and ideological deep roots .
After that, it seems that philosophers who assert that values are absolute reflect conservative ideology , and want to fix social reality to serve their concerns; while philosophers who assert that values are relative reflect progressive ideology, and want to change social reality .
This research also clarifies that philosophers and thinkers of absolute values believe in absolute facts and knowledge; while philosophers and thinkers of relative values believe in relative facts and knowledge.
