While much research has been done in text-to-image synthesis, little work has been done to explore the usage of linguistic structure of the input text. Such information is even more important for story visualization since its inputs have an explicit
narrative structure that needs to be translated into an image sequence (or visual story). Prior work in this domain has shown that there is ample room for improvement in the generated image sequence in terms of visual quality, consistency and relevance. In this paper, we first explore the use of constituency parse trees using a Transformer-based recurrent architecture for encoding structured input. Second, we augment the structured input with commonsense information and study the impact of this external knowledge on the generation of visual story. Third, we also incorporate visual structure via bounding boxes and dense captioning to provide feedback about the characters/objects in generated images within a dual learning setup. We show that off-the-shelf dense-captioning models trained on Visual Genome can improve the spatial structure of images from a different target domain without needing fine-tuning. We train the model end-to-end using intra-story contrastive loss (between words and image sub-regions) and show significant improvements in visual quality. Finally, we provide an analysis of the linguistic and visuo-spatial information.
Stance detection (SD) entails classifying the sentiment of a text towards a given target, and is a relevant sub-task for opinion mining and social media analysis. Recent works have explored knowledge infusion supplementing the linguistic competence a
nd latent knowledge of large pre-trained language models with structured knowledge graphs (KGs), yet few works have applied such methods to the SD task. In this work, we first perform stance-relevant knowledge probing on Transformers-based pre-trained models in a zero-shot setting, showing these models' latent real-world knowledge about SD targets and their sensitivity to context. We then train and evaluate new knowledge-enriched stance detection models on two Twitter stance datasets, achieving state-of-the-art performance on both.
For most language combinations and parallel data is either scarce or simply unavailable. To address this and unsupervised machine translation (UMT) exploits large amounts of monolingual data by using synthetic data generation techniques such as back-
translation and noising and while self-supervised NMT (SSNMT) identifies parallel sentences in smaller comparable data and trains on them. To this date and the inclusion of UMT data generation techniques in SSNMT has not been investigated. We show that including UMT techniques into SSNMT significantly outperforms SSNMT (up to +4.3 BLEU and af2en) as well as statistical (+50.8 BLEU) and hybrid UMT (+51.5 BLEU) baselines on related and distantly-related and unrelated language pairs.
Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to ex
tend an arbitrary set of relational database tables with information extracted from a document corpus. OpenKi[1] addresses this task through extraction of named entities and predicates via OpenIE tools then learning relation embeddings from the resulting entity-relation graph for relation prediction, outperforming previous approaches. We present an extension of OpenKi that incorporates embeddings of text-based representations of the entities and the relations. We demonstrate that this results in a substantial performance increase over a system without this information.
استخدام ملفات الفيديو للتغلب على المشاكل الناتجة عن اللغة الأم و التعميم الخاطئ للمفردات في تعليم المفردات: دراسة أجريت في مؤسسة ميغاسوفت للتدريب
Studing and defining the types of soils by using the method of
integrating data of remote sensing and the devices measuring the
reflex rays such as radiometer and spectrometer ….etc is considered
one of very recent technology in such study,so that
we define the
reflexed spectrum intensity for the ground targets where the soil is
one of them that directly in the field and making comparsion of this
results with images of satellites where the channels of device do on
the same spectrum ranges of the satellites.
The expense of using the ground method is high an it requires
long time comparativilty using the space images to the same target.
In this research types of soils determined their speed
boundaries,drawing a map and making the reflexed graph for them.
Worldwide, Information and communication technologies (ICTs)
have gained a vital importance in satisfying the requirements of both the
new Education System and the revised Curriculum for English Language
Teaching (ELT). The experience of introduci
ng different ICTs into the
classroom and other educational settings worldwide suggests that
teachers' effective integration of technology into their curricula largely
depends on their perceptions and attitudes to the value of these tools to
achieving their instructional purposes. This research seeks to evaluate the
readiness to integrate ICTs into the current teaching practices of the
faculty at the department of English Language and Literature at
Damascus University. In a context of large classes where resources are
minimal and staff members are in short supply, using technology assisted
learning at Damascus University may help enhance education by giving
learners more flexibility and helping them evolve into more autonomous
learners. Investigating the current faculty’s reading of the value of
technology as a means of overcoming some of the restrictions imposed
by their context will uncover potential limitations on the use of ICTs in
the existing context and suggest workable solutions.