Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userLanguage Resource Efficient Learning for Captioning
تعلم اللغة التعلم كفاءة في الموارد
263
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
نظرا للجهود المعجدة والإدراكية المتورطة في التوليد اليدوي من تعليق واحد لكل إدخال الصورة / الفيديو، فإن موارد الشروح البشرية محدودة للغاية لتسمية المهام. نحن نحدد كفاءة الموارد اللغوية باعتبارها تصل إلى نفس الأداء مع التوضيحية المشروحة أقل لكل مدخلات. ندرس أولا تدهور أداء نماذج التسمية التوضيحية في إعدادات موارد لغة مختلفة. يظهر تحليل نماذج التسمية التوضيحية مع خسارة SC أن تدهور الأداء ناتج عن تقدير المكافأة بشكل متزايد للمكافآت والأساس مع عدد أقل من الموارد اللغوية. لتخفيف هذه المشكلة، نقترح تقليل تباين الضوضاء في الأساس عن طريق تعميم المقارنة الزوجية الفردية في فقدان SC واستخدام مقارنات الزوجية المتعددة المتعمدة. يقيس المقارنة الزوجية المعممة (GPC) الفرق بين درجات التقييم التوضيحتين فيما يتعلق بالإدخال. وإظهار تجريبيا، نظرا لأن النموذج الذي تم تدريبه مع فقدان GPC المقترح فعالا على مورد اللغة وتحقق أداء مماثل مع النماذج الحديثة على MSCOC باستخدام نصف موارد اللغة فقط. علاوة على ذلك، تتفوق نموذجنا بشكل كبير على النماذج الحديثة على مجموعة بيانات تعليق الفيديو التي تحتوي على علامة تعليق واحدة فقط لكل إدخال في مجموعة التدريب.
Due to complex cognitive and inferential efforts involved in the manual generation of one caption per image/video input, the human annotation resources are very limited for captioning tasks. We define language resource efficient as reaching the same performance with fewer annotated captions per input. We first study the performance degradation of caption models in different language resource settings. Our analysis of caption models with SC loss shows that the performance degradation is caused by the increasingly noisy estimation of reward and baseline with fewer language resources. To mitigate this issue, we propose to reduce the variance of noise in the baseline by generalizing the single pairwise comparison in SC loss and using multiple generalized pairwise comparisons. The generalized pairwise comparison (GPC) measures the difference between the evaluation scores of two captions with respect to an input. Empirically, we show that the model trained with the proposed GPC loss is efficient on language resource and achieves similar performance with the state-of-the-art models on MSCOCO by using only half of the language resources. Furthermore, our model significantly outperforms the state-of-the-art models on a video caption dataset that has only one labeled caption per input in the training set.
References used
https://aclanthology.org/
rate research
Read More
Meta-learning has achieved great success in leveraging the historical learned knowledge to facilitate the learning process of the new task. However, merely learning the knowledge from the historical tasks, adopted by current meta-learning algorithms,
may not generalize well to testing tasks when they are not well-supported by training tasks. This paper studies a low-resource text classification problem and bridges the gap between meta-training and meta-testing tasks by leveraging the external knowledge bases. Specifically, we propose KGML to introduce additional representation for each sentence learned from the extracted sentence-specific knowledge graph. The extensive experiments on three datasets demonstrate the effectiveness of KGML under both supervised adaptation and unsupervised adaptation settings.
Open and distance learning is experiencing a rapid growth
throughout the world and Syria is no exception. With the turn of the new
millennium, Syria launched two state institutes for distance learning: The
Open Learning Centre (opened in 2001) and
the Syrian Virtual University
(opened in 2002). The Syrian Virtual University (SVU), one of its kind in
the whole Arab region, offers students the opportunity to gain education
through an online learning environment based on the latest technology.
Since Syria is a country where English has become an important
educational requirement, the teaching of English as a foreign language
has therefore entered the arena of distance learning.
News recommendation is critical for personalized news access. Most existing news recommendation methods rely on centralized storage of users' historical news click behavior data, which may lead to privacy concerns and hazards. Federated Learning is a
privacy-preserving framework for multiple clients to collaboratively train models without sharing their private data. However, the computation and communication cost of directly learning many existing news recommendation models in a federated way are unacceptable for user clients. In this paper, we propose an efficient federated learning framework for privacy-preserving news recommendation. Instead of training and communicating the whole model, we decompose the news recommendation model into a large news model maintained in the server and a light-weight user model shared on both server and clients, where news representations and user model are communicated between server and clients. More specifically, the clients request the user model and news representations from the server, and send their locally computed gradients to the server for aggregation. The server updates its global user model with the aggregated gradients, and further updates its news model to infer updated news representations. Since the local gradients may contain private information, we propose a secure aggregation method to aggregate gradients in a privacy-preserving way. Experiments on two real-world datasets show that our method can reduce the computation and communication cost on clients while keep promising model performance.
This paper describes our submission for the shared task on Unsupervised MT and Very Low Resource Supervised MT at WMT 2021. We submitted systems for two language pairs: German ↔ Upper Sorbian (de ↔ hsb) and German-Lower Sorbian (de ↔ dsb). For de ↔ h
sb, we pretrain our system using MASS (Masked Sequence to Sequence) objective and then finetune using iterative back-translation. Final finetunng is performed using the parallel data provided for translation objective. For de ↔ dsb, no parallel data is provided in the task, we use final de ↔ hsb model as initialization of the de ↔ dsb model and train it further using iterative back-translation, using the same vocabulary as used in the de ↔ hsb model.
Adapters are light-weight modules that allow parameter-efficient fine-tuning of pretrained models. Specialized language and task adapters have recently been proposed to facilitate cross-lingual transfer of multilingual pretrained models (Pfeiffer et
al., 2020b). However, this approach requires training a separate language adapter for every language one wishes to support, which can be impractical for languages with limited data. An intuitive solution is to use a related language adapter for the new language variety, but we observe that this solution can lead to sub-optimal performance. In this paper, we aim to improve the robustness of language adapters to uncovered languages without training new adapters. We find that ensembling multiple existing language adapters makes the fine-tuned model significantly more robust to other language varieties not included in these adapters. Building upon this observation, we propose Entropy Minimized Ensemble of Adapters (EMEA), a method that optimizes the ensemble weights of the pretrained language adapters for each test sentence by minimizing the entropy of its predictions. Experiments on three diverse groups of language varieties show that our method leads to significant improvements on both named entity recognition and part-of-speech tagging across all languages.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
