Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userEfficient Test Time Adapter Ensembling for Low-resource Language Varieties
الفعال محول اختبار الوقت الكفاءة بالنسبة لأصناف لغة الموارد المنخفضة
263
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
المحولات هي وحدات خفيفة الوزن تسمح بضبط النماذج الدقيقة التي يتمتع بها المعلمة. تم اقتراح محولات اللغة والمهمة المتخصصة مؤخرا لتسهيل التحويل عبر اللغات للنماذج المحددة متعددة اللغات (Pfeiffer et al.، 2020b). ومع ذلك، يتطلب هذا النهج تدريب محول لغة منفصل لكل لغة ترغب في الدعم، مما قد يكون غير صحيح لغات مع بيانات محدودة. الحل البديهي هو استخدام محول لغة ذات صلة لتنوع اللغات الجديدة، لكننا نلاحظ أن هذا الحل يمكن أن يؤدي إلى الأداء دون الأمثل. في هذه الورقة، نهدف إلى تحسين متانة المحولات اللغوية باللغات غير المكشوفة دون تدريب محولات جديدة. نجد أن الكشف عن محولات متعددة اللغات متعددة يجعل النموذج الدقيق أكثر قوة أكثر بكثير من أصناف اللغة الأخرى غير المدرجة في هذه المحولات. بناء على هذه الملاحظة، نقترح Entropy Minimized Entermble of Adrapters (EMEA)، وهي طريقة تعمل على تحسين أوزان مجموعة محولات اللغة المحددة مسبقا لكل جملة اختبار عن طريق تقليل انتروبيا من تنبؤاتها. تبين التجارب في ثلاث مجموعات متنوعة من الأصناف اللغوية أن طريقتنا تؤدي إلى تحسينات كبيرة على كل من الاعتراف الكياري المسمى ووضع علامات جزء من الكلام في جميع اللغات.
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.
References used
https://aclanthology.org/
rate research
Read More
In this paper, we develop Sindhi subjective lexicon using a merger of existing English resources: NRC lexicon, list of opinion words, SentiWordNet, Sindhi-English bilingual dictionary, and collection of Sindhi modifiers. The positive or negative sent
iment score is assigned to each Sindhi opinion word. Afterwards, we determine the coverage of the proposed lexicon with subjectivity analysis. Moreover, we crawl multi-domain tweet corpus of news, sports, and finance. The crawled corpus is annotated by experienced annotators using the Doccano text annotation tool. The sentiment annotated corpus is evaluated by employing support vector machine (SVM), recurrent neural network (RNN) variants, and convolutional neural network (CNN).
A bigger is better'' explosion in the number of parameters in deep neural networks has made it increasingly challenging to make state-of-the-art networks accessible in compute-restricted environments. Compression techniques have taken on renewed impo
rtance as a way to bridge the gap. However, evaluation of the trade-offs incurred by popular compression techniques has been centered on high-resource datasets. In this work, we instead consider the impact of compression in a data-limited regime. We introduce the term low-resource double bind to refer to the co-occurrence of data limitations and compute resource constraints. This is a common setting for NLP for low-resource languages, yet the trade-offs in performance are poorly studied. Our work offers surprising insights into the relationship between capacity and generalization in data-limited regimes for the task of machine translation. Our experiments on magnitude pruning for translations from English into Yoruba, Hausa, Igbo and German show that in low-resource regimes, sparsity preserves performance on frequent sentences but has a disparate impact on infrequent ones. However, it improves robustness to out-of-distribution shifts, especially for datasets that are very distinct from the training distribution. Our findings suggest that sparsity can play a beneficial role at curbing memorization of low frequency attributes, and therefore offers a promising solution to the low-resource double bind.
Adapter modules have emerged as a general parameter-efficient means to specialize a pretrained encoder to new domains. Massively multilingual transformers (MMTs) have particularly benefited from additional training of language-specific adapters. Howe
ver, this approach is not viable for the vast majority of languages, due to limitations in their corpus size or compute budgets. In this work, we propose MAD-G (Multilingual ADapter Generation), which contextually generates language adapters from language representations based on typological features. In contrast to prior work, our time- and space-efficient MAD-G approach enables (1) sharing of linguistic knowledge across languages and (2) zero-shot inference by generating language adapters for unseen languages. We thoroughly evaluate MAD-G in zero-shot cross-lingual transfer on part-of-speech tagging, dependency parsing, and named entity recognition. While offering (1) improved fine-tuning efficiency (by a factor of around 50 in our experiments), (2) a smaller parameter budget, and (3) increased language coverage, MAD-G remains competitive with more expensive methods for language-specific adapter training across the board. Moreover, it offers substantial benefits for low-resource languages, particularly on the NER task in low-resource African languages. Finally, we demonstrate that MAD-G's transfer performance can be further improved via: (i) multi-source training, i.e., by generating and combining adapters of multiple languages with available task-specific training data; and (ii) by further fine-tuning generated MAD-G adapters for languages with monolingual data.
This paper describes the participation of team oneNLP (LTRC, IIIT-Hyderabad) for the WMT 2021 task, similar language translation. We experimented with transformer based Neural Machine Translation and explored the use of language similarity for Tamil-
Telugu and Telugu-Tamil. We incorporated use of different subword configurations, script conversion and single model training for both directions as exploratory experiments.
The widespread presence of offensive language on social media motivated the development of systems capable of recognizing such content automatically. Apart from a few notable exceptions, most research on automatic offensive language identification ha
s dealt with English. To address this shortcoming, we introduce MOLD, the Marathi Offensive Language Dataset. MOLD is the first dataset of its kind compiled for Marathi, thus opening a new domain for research in low-resource Indo-Aryan languages. We present results from several machine learning experiments on this dataset, including zero-short and other transfer learning experiments on state-of-the-art cross-lingual transformers from existing data in Bengali, English, and Hindi.
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
