Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userPushing the Right Buttons: Adversarial Evaluation of Quality Estimation
دفع الأزرار المناسبة: تقييم الخصم لتقدير الجودة
133
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
توفر أنظمة ترجمة الآلات الحالية (MT) نتائج جيدة للغاية على مجموعة متنوعة متنوعة من أزواج اللغات ومجموعات البيانات. ومع ذلك، من المعروف أن إنتاج مخرجات ترجمة بطلاقة يمكن أن تحتوي على أخطاء ذات معنى مهم، وبالتالي تقويض موثوقيتها في الممارسة العملية. تقدير الجودة (QE) هي مهمة تقييم أداء MT تلقائيا في وقت الاختبار. وبالتالي، من أجل أن تكون مفيدة، يجب أن تكون أنظمة QE قادرا على اكتشاف هذه الأخطاء. ومع ذلك، لم يتم بعد اختبار هذه القدرة في ممارسات التقييم الحالية، حيث يتم تقييم أنظمة QE فقط من حيث ارتباطها بالأحكام البشرية. في هذا العمل، نسرج هذه الفجوة من خلال اقتراح منهجية عامة لاختبار الخصم من QE ل MT. أولا، نظرا لأنه على الرغم من الارتباط العالي بالأحكام البشرية التي حققتها سوتا الأخيرة، لا تزال أنواع معينة من الأخطاء المعنية مشكلة في الكشف عن QE. ثانيا، نظهر أنه في المتوسط، فإن قدرة نموذج معين على التمييز بين الاضطرابات المعنية التي تحافظ مع المعنى وتغيير المعنى هي التنبؤ بأدائها العام، وبالتالي يحتمل أن يسمح بمقارنة أنظمة QE دون الاعتماد على تشريح الجودة اليدوية.
Current Machine Translation (MT) systems achieve very good results on a growing variety of language pairs and datasets. However, they are known to produce fluent translation outputs that can contain important meaning errors, thus undermining their reliability in practice. Quality Estimation (QE) is the task of automatically assessing the performance of MT systems at test time. Thus, in order to be useful, QE systems should be able to detect such errors. However, this ability is yet to be tested in the current evaluation practices, where QE systems are assessed only in terms of their correlation with human judgements. In this work, we bridge this gap by proposing a general methodology for adversarial testing of QE for MT. First, we show that despite a high correlation with human judgements achieved by the recent SOTA, certain types of meaning errors are still problematic for QE to detect. Second, we show that on average, the ability of a given model to discriminate between meaning-preserving and meaning-altering perturbations is predictive of its overall performance, thus potentially allowing for comparing QE systems without relying on manual quality annotation.
References used
https://aclanthology.org/
rate research
Read More
We propose a novel scheme to use the Levenshtein Transformer to perform the task of word-level quality estimation. A Levenshtein Transformer is a natural fit for this task: trained to perform decoding in an iterative manner, a Levenshtein Transformer
can learn to post-edit without explicit supervision. To further minimize the mismatch between the translation task and the word-level QE task, we propose a two-stage transfer learning procedure on both augmented data and human post-editing data. We also propose heuristics to construct reference labels that are compatible with subword-level finetuning and inference. Results on WMT 2020 QE shared task dataset show that our proposed method has superior data efficiency under the data-constrained setting and competitive performance under the unconstrained setting.
Most current quality estimation (QE) models for machine translation are trained and evaluated in a fully supervised setting requiring significant quantities of labelled training data. However, obtaining labelled data can be both expensive and time-co
nsuming. In addition, the test data that a deployed QE model would be exposed to may differ from its training data in significant ways. In particular, training samples are often labelled by one or a small set of annotators, whose perceptions of translation quality and needs may differ substantially from those of end-users, who will employ predictions in practice. Thus, it is desirable to be able to adapt QE models efficiently to new user data with limited supervision data. To address these challenges, we propose a Bayesian meta-learning approach for adapting QE models to the needs and preferences of each user with limited supervision. To enhance performance, we further propose an extension to a state-of-the-art Bayesian meta-learning approach which utilizes a matrix-valued kernel for Bayesian meta-learning of quality estimation. Experiments on data with varying number of users and language characteristics demonstrates that the proposed Bayesian meta-learning approach delivers improved predictive performance in both limited and full supervision settings.
In quality estimation (QE), the quality of translation can be predicted by referencing the source sentence and the machine translation (MT) output without access to the reference sentence. However, there exists a paradox in that constructing a datase
t for creating a QE model requires non-trivial human labor and time, and it may even requires additional effort compared to the cost of constructing a parallel corpus. In this study, to address this paradox and utilize the various applications of QE, even in low-resource languages (LRLs), we propose a method for automatically constructing a pseudo-QE dataset without using human labor. We perform a comparative analysis on the pseudo-QE dataset using multilingual pre-trained language models. As we generate the pseudo dataset, we conduct experiments using various external machine translators as test sets to verify the accuracy of the results objectively. Also, the experimental results show that multilingual BART demonstrates the best performance, and we confirm the applicability of QE in LRLs using pseudo-QE dataset construction methods.
This paper describes Papago submission to the WMT 2021 Quality Estimation Task 1: Sentence-level Direct Assessment. Our multilingual Quality Estimation system explores the combination of Pretrained Language Models and Multi-task Learning architecture
s. We propose an iterative training pipeline based on pretraining with large amounts of in-domain synthetic data and finetuning with gold (labeled) data. We then compress our system via knowledge distillation in order to reduce parameters yet maintain strong performance. Our submitted multilingual systems perform competitively in multilingual and all 11 individual language pair settings including zero-shot.
In this shared task, we seek the participating teams to investigate the factors influencing the quality of the code-mixed text generation systems. We synthetically generate code-mixed Hinglish sentences using two distinct approaches and employ human
annotators to rate the generation quality. We propose two subtasks, quality rating prediction and annotators' disagreement prediction of the synthetic Hinglish dataset. The proposed subtasks will put forward the reasoning and explanation of the factors influencing the quality and human perception of the code-mixed text.
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
