الإعلان عن التجارة الإلكترونية ومواقع وسائل التواصل الاجتماعي تقدم انطباعات إعلانية على نطاق الويب بقيمة قيادة يومية لكل من المتسوقين والمعلنين.يتطلب هذا المقياس طرقا برنامجية للكشف عن محتوى غير مناسب في الإعلانات لحماية تجربة العملاء والثقة.يركز هذه الورق على تقنيات تدريب نماذج تصنيف النص بموجب قيود الموارد، بنيت كجزء من الحلول الآلية للاعتدال محتوى الإعلان.نظير على مدى ضعف الإشراف، يمكن تطبيق تعلم المناهج الدراسية والتدريب متعدد اللغات بفعالية على Tune-Tune Bert ومتغيراته لمهام تصنيف النص بالاقتران مع مختلف استراتيجيات تكبير البيانات.تبين تجاربنا الواسعة لغات متعددة أن هذه التقنيات تكتشف فئات الإعلانات العدائية مع مكاسب كبيرة في الدقة في عتبة استدعاء عالية فوق الأساس.
Advertising on e-commerce and social media sites deliver ad impressions at web scale on a daily basis driving value to both shoppers and advertisers. This scale necessitates programmatic ways of detecting unsuitable content in ads to safeguard customer experience and trust. This paper focusses on techniques for training text classification models under resource constraints, built as part of automated solutions for advertising content moderation. We show how weak supervision, curriculum learning and multi-lingual training can be applied effectively to fine-tune BERT and its variants for text classification tasks in conjunction with different data augmentation strategies. Our extensive experiments on multiple languages show that these techniques detect adversarial ad categories with a substantial gain in precision at high recall threshold over the baseline.
References used
https://aclanthology.org/
Cross-target generalization constitutes an important issue for news Stance Detection (SD). In this short paper, we investigate adversarial cross-genre SD, where knowledge from annotated user-generated data is leveraged to improve news SD on targets u
Large language models have become increasingly difficult to train because of the growing computation time and cost. In this work, we present SRU++, a highly-efficient architecture that combines fast recurrence and attention for sequence modeling. SRU
Detecting stance on Twitter is especially challenging because of the short length of each tweet, the continuous coinage of new terminology and hashtags, and the deviation of sentence structure from standard prose. Fine-tuned language models using lar
The problem of detecting psychological stress in online posts, and more broadly, of detecting people in distress or in need of help, is a sensitive application for which the ability to interpret models is vital. Here, we present work exploring the us
We present a simple yet effective Targeted Adversarial Training (TAT) algorithm to improve adversarial training for natural language understanding. The key idea is to introspect current mistakes and prioritize adversarial training steps to where the