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.
Position representation is crucial for building position-aware representations in Transformers. Existing position representations suffer from a lack of generalization to test data with unseen lengths or high computational cost. We investigate shifted
absolute position embedding (SHAPE) to address both issues. The basic idea of SHAPE is to achieve shift invariance, which is a key property of recent successful position representations, by randomly shifting absolute positions during training. We demonstrate that SHAPE is empirically comparable to its counterpart while being simpler and faster.
As NLP systems become better at detecting opinions and beliefs from text, it is important to ensure not only that models are accurate but also that they arrive at their predictions in ways that align with human reasoning. In this work, we present a m
ethod for imparting human-like rationalization to a stance detection model using crowdsourced annotations on a small fraction of the training data. We show that in a data-scarce setting, our approach can improve the reasoning of a state-of-the-art classifier---particularly for inputs containing challenging phenomena such as sarcasm---at no cost in predictive performance. Furthermore, we demonstrate that attention weights surpass a leading attribution method in providing faithful explanations of our model's predictions, thus serving as a computationally cheap and reliable source of attributions for our model.
Abuse on the Internet is an important societal problem of our time. Millions of Internet users face harassment, racism, personal attacks, and other types of abuse across various platforms. The psychological effects of abuse on individuals can be prof
ound and lasting. Consequently, over the past few years, there has been a substantial research effort towards automated abusive language detection in the field of NLP. In this position paper, we discuss the role that modeling of users and online communities plays in abuse detection. Specifically, we review and analyze the state of the art methods that leverage user or community information to enhance the understanding and detection of abusive language. We then explore the ethical challenges of incorporating user and community information, laying out considerations to guide future research. Finally, we address the topic of explainability in abusive language detection, proposing properties that an explainable method should aim to exhibit. We describe how user and community information can facilitate the realization of these properties and discuss the effective operationalization of explainability in view of the properties.
Quality estimation (QE) of machine translation (MT) aims to evaluate the quality of machine-translated sentences without references and is important in practical applications of MT. Training QE models require massive parallel data with hand-crafted q
uality annotations, which are time-consuming and labor-intensive to obtain. To address the issue of the absence of annotated training data, previous studies attempt to develop unsupervised QE methods. However, very few of them can be applied to both sentence- and word-level QE tasks, and they may suffer from noises in the synthetic data. To reduce the negative impact of noises, we propose a self-supervised method for both sentence- and word-level QE, which performs quality estimation by recovering the masked target words. Experimental results show that our method outperforms previous unsupervised methods on several QE tasks in different language pairs and domains.
Stance detection on social media can help to identify and understand slanted news or commentary in everyday life. In this work, we propose a new model for zero-shot stance detection on Twitter that uses adversarial learning to generalize across topic
s. Our model achieves state-of-the-art performance on a number of unseen test topics with minimal computational costs. In addition, we extend zero-shot stance detection to topics not previously considered, highlighting future directions for zero-shot transfer.
تناولت في هذا البحث الموجز نماذج من شواذ التصريف , فوقفت على بعض آراء النحاة المتقدمين و اللاحقين و بينت تناولهم لظاهرة الشذوذ, و كيف جعلوا لها قواعد عامة تنظمها , و قد آثرت أن استعرض -مع المناقشة- بعض الألفاظ التي شذت عن القاعدة, و التي كانت مدار خل
اف بين النحاة , كبعض شواذ الإعلال و التصغير , و بينت وجهة نظري فيها , ثم انتهيت إلى نتائج دونتها في نهاية البحث.
The study addresses a number of legal issues related to the Syrian position on the
Special Tribunal for Lebanon. First, it represents the legal basis of the Syria position
with regard to the rules of the national law.
The study also addresses a nu
mber of issues and legal questions to the extent of
compliance of this position with international law, and precidents related to the
creation of former international tribunals. It also looks into the issue of consistency
between the Syrian position and the special tribunal’s statute. It raises the question,
based on legal documents issued by the tribunal, on countries’ obligations, including
Syria, to future cooperation with the Tribunal on issues of opening investigations,
extradition and the renunciation of jurisdiction. The study addresses the impact and
legal consequences resulting from these issues through highlighting the tribunal’s
position on these issues and questions.
يتناول بحثي موقف الإمام الغزالي من علم الكلام و أدلة المتكلمين، فقد اختلف العلماء في أهمية علم الكلام و حاجة الناس إليه، و كان الغزالي يرى أن هذا العلم ليس له إلا مهمة واحدة تتجلى في الرد على الشبهات و حفظ عقائد العوام، فعلم الكلام دواء لا غذاء و الد
واء لا يحتاجه إلا المريض، كما تكلم عن ضرره و فائدته فقال بأن ضرره يتجلى في إثارة الشبهات و في تأكيد اعتقاد المبتدعة، فالهوى و التعصب و بغض الخصوم يستولي على قلب المتكلم و يمنعه من إدراك الحق أو التسليم له، و يتضمن مجادلة مذمومة كما يقوم على المشاغبة بالتعلق بمناقضات الفرق. أما منفعته فتظهر في نوع واحد و هو حراسة العقيدة على العوام و حفظها عن تشويشات المبتدعة بأنواع الجدل، فإن العامي ضعيف يستفزه جدل المبتدع و إن كان فاسداً و معارضة الفاسد بالفاسد تدفعه، لذلك نجد أنه على حكم على علم الكلام عدة أحكام مختلفة، فهو واجب كفائي، و حرام، و واجب عيني. فمرة يصرح بالتحريم لشدة ضرره على عوام الخلق و كثرة آفاته، و هو واجب عيني في حالة خاصة لشخص تمكنت البدعة منه و لم يستطع أن يزيلها و يعيد الطمأنينة لنفسه إلا بممارسة نوع من
الكلام، و هو واجب كفائي لقمع المبتدعة و حراسة العقيدة و هي مهمة دفاعية لا يحتاج إليها سائر الخلق..
اختلفت التشريعات في موقفها من أساس المسؤولية المدنية التقصيرية، فبعضها
أقامها على أساس الخطأ في حين أقامها بعضهم الآخر على أساس الفعل الضار، و هدف هذا
البحث هو بيان الفعل الضار و أركان المسؤولية المدنية التقصيرية عامة و في زراعة التبغ
و صناعته و ت
دخين السجائر خاصة، و تحديد القانون الواجب تطبيقه عليها و المحكمة المختصة
برؤيتها في حالة تنازع القوانين بشأنها في ضوء دعاوي التعويض التي أقيمت مؤخرًا على
شركات صنع السجائر في بعض الدول الأوروبية.
