Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userTowards Developing a Multilingual and Code-Mixed Visual Question Answering System by Knowledge Distillation
نحو تطوير سؤال مرئي متعدد اللغات ومزوج التعليمات البرمجية من خلال تنطير المعرفة
219
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
أظهرت نماذج الرؤية اللغوية المدربة مسبقا أداء رائعا حول مهمة الإجابة على السؤال المرئي (VQA). ومع ذلك، يتم تدريب معظم النماذج المدربة مسبقا من خلال النظر فقط في التعلم أحادي الأونلينغ، وخاصة اللغة الغنية بالموارد مثل اللغة الإنجليزية. تدريب هذه النماذج للكمات متعددة اللغات طلب موارد الحوسبة عالية ومجموعات بيانات الرؤية متعددة اللغات التي تعيق تطبيقها في الممارسة العملية. لتخفيف هذه التحديات، نقترح نهج تقطير المعرفة لتوسيع نموذج للرؤية باللغة الإنجليزية (المعلم) في نموذج متعدد اللغات ومزوج التعليمات البرمجية (طالبة). على عكس أساليب تقطير المعرفة الحالية، والتي تستخدم فقط الإخراج من الطبقة الأخيرة من شبكة المعلم للتقطير، يتعلم نموذج الطالب الخاص بنا وتقليد المعلم من طبقات متعددة الوسائط (تشفير اللغة والرؤية) بأهداف تقطير مصممة بشكل مناسب لاستخراج المعرفة الإضافية وبعد كما نقوم بإنشاء مجموعة بيانات VQA متعددة اللغات متعددة اللغات متعددة اللغات وخلطها في أحد عشر جهازا مختلفا للنظر في اللغات الهندية والأوروبية المتعددة. تظهر النتائج التجريبية والتحليل المتعمق فعالية نموذج VQA المقترح على نماذج الرؤية المدربة مسبقا في الرؤية المدربة مسبقا في أحد عشر من إعدادات لغة متنوعة.
Pre-trained language-vision models have shown remarkable performance on the visual question answering (VQA) task. However, most pre-trained models are trained by only considering monolingual learning, especially the resource-rich language like English. Training such models for multilingual setups demand high computing resources and multilingual language-vision dataset which hinders their application in practice. To alleviate these challenges, we propose a knowledge distillation approach to extend an English language-vision model (teacher) into an equally effective multilingual and code-mixed model (student). Unlike the existing knowledge distillation methods, which only use the output from the last layer of the teacher network for distillation, our student model learns and imitates the teacher from multiple intermediate layers (language and vision encoders) with appropriately designed distillation objectives for incremental knowledge extraction. We also create the large-scale multilingual and code-mixed VQA dataset in eleven different language setups considering the multiple Indian and European languages. Experimental results and in-depth analysis show the effectiveness of the proposed VQA model over the pre-trained language-vision models on eleven diverse language setups.
References used
https://aclanthology.org/
rate research
Read More
Previous existing visual question answering (VQA) systems commonly use graph neural networks(GNNs) to extract visual relationships such as semantic relations or spatial relations. However, studies that use GNNs typically ignore the importance of each
relation and simply concatenate outputs from multiple relation encoders. In this paper, we propose a novel layer architecture that fuses multiple visual relations through an attention mechanism to address this issue. Specifically, we develop a model that uses question embedding and joint embedding of the encoders to obtain dynamic attention weights with regard to the type of questions. Using the learnable attention weights, the proposed model can efficiently use the necessary visual relation features for a given question. Experimental results on the VQA 2.0 dataset demonstrate that the proposed model outperforms existing graph attention network-based architectures. Additionally, we visualize the attention weight and show that the proposed model assigns a higher weight to relations that are more relevant to the question.
We study multilingual AMR parsing from the perspective of knowledge distillation, where the aim is to learn and improve a multilingual AMR parser by using an existing English parser as its teacher. We constrain our exploration in a strict multilingua
l setting: there is but one model to parse all different languages including English. We identify that noisy input and precise output are the key to successful distillation. Together with extensive pre-training, we obtain an AMR parser whose performances surpass all previously published results on four different foreign languages, including German, Spanish, Italian, and Chinese, by large margins (up to 18.8 Smatch points on Chinese and on average 11.3 Smatch points). Our parser also achieves comparable performance on English to the latest state-of-the-art English-only parser.
Text generation is a highly active area of research in the computational linguistic community. The evaluation of the generated text is a challenging task and multiple theories and metrics have been proposed over the years. Unfortunately, text generat
ion and evaluation are relatively understudied due to the scarcity of high-quality resources in code-mixed languages where the words and phrases from multiple languages are mixed in a single utterance of text and speech. To address this challenge, we present a corpus (HinGE) for a widely popular code-mixed language Hinglish (code-mixing of Hindi and English languages). HinGE has Hinglish sentences generated by humans as well as two rule-based algorithms corresponding to the parallel Hindi-English sentences. In addition, we demonstrate the in- efficacy of widely-used evaluation metrics on the code-mixed data. The HinGE dataset will facilitate the progress of natural language generation research in code-mixed languages.
A major challenge in analysing social me-dia data belonging to languages that use non-English script is its code-mixed nature. Recentresearch has presented state-of-the-art contex-tual embedding models (both monolingual s.a.BERT and multilingual s.a.
XLM-R) as apromising approach. In this paper, we showthat the performance of such embedding mod-els depends on multiple factors, such as thelevel of code-mixing in the dataset, and thesize of the training dataset. We empiricallyshow that a newly introduced Capsule+biGRUclassifier could outperform a classifier built onthe English-BERT as well as XLM-R just witha training dataset of about 6500 samples forthe Sinhala-English code-mixed data.
We tackle multi-choice question answering. Acquiring related commonsense knowledge to the question and options facilitates the recognition of the correct answer. However, the current reasoning models suffer from the noises in the retrieved knowledge.
In this paper, we propose a novel encoding method which is able to conduct interception and soft filtering. This contributes to the harvesting and absorption of representative information with less interference from noises. We experiment on CommonsenseQA. Experimental results illustrate that our method yields substantial and consistent improvements compared to the strong Bert, RoBERTa and Albert-based baselines.
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
