اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدLearning Feature Weights using Reward Modeling for Denoising Parallel Corpora
71
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Large web-crawled corpora represent an excellent resource for improving the performance of Neural Machine Translation (NMT) systems across several language pairs. However, since these corpora are typically extremely noisy, their use is fairly limited. Current approaches to dealing with this problem mainly focus on filtering using heuristics or single features such as language model scores or bi-lingual similarity. This work presents an alternative approach which learns weights for multiple sentence-level features. These feature weights which are optimized directly for the task of improving translation performance, are used to score and filter sentences in the noisy corpora more effectively. We provide results of applying this technique to building NMT systems using the Paracrawl corpus for Estonian-English and show that it beats strong single feature baselines and hand designed combinations. Additionally, we analyze the sensitivity of this method to different types of noise and explore if the learned weights generalize to other language pairs using the Maltese-English Paracrawl corpus.
قيم البحث
اقرأ أيضاً
We propose a new approach for learning contextualised cross-lingual word embeddings based only on a small parallel corpus (e.g. a few hundred sentence pairs). Our method obtains word embeddings via an LSTM-based encoder-decoder model that performs bi
directional translation and reconstruction of the input sentence. Through sharing model parameters among different languages, our model jointly trains the word embeddings in a common multilingual space. We also propose a simple method to combine word and subword embeddings to make use of orthographic similarities across different languages. We base our experiments on real-world data from endangered languages, namely Yongning Na, Shipibo-Konibo and Griko. Our experiments on bilingual lexicon induction and word alignment tasks show that our model outperforms existing methods by a large margin for most language pairs. These results demonstrate that, contrary to common belief, an encoder-decoder translation model is beneficial for learning cross-lingual representations, even in extremely low-resource scenarios.
Journalists obtain leads, or story ideas, by reading large corpora of government records: court cases, proposed bills, etc. However, only a small percentage of such records are interesting documents. We propose a model of newsworthiness aimed at surf
acing interesting documents. We train models on automatically labeled corpora -- published newspaper articles -- to predict whether each article was a front-page article (i.e., textbf{newsworthy}) or not (i.e., textbf{less newsworthy}). We transfer these models to unlabeled corpora -- court cases, bills, city-council meeting minutes -- to rank documents in these corpora on newsworthiness. A fine-tuned RoBERTa model achieves .93 AUC performance on heldout labeled documents, and .88 AUC on expert-validated unlabeled corpora. We provide interpretation and visualization for our models.
We apply rule induction, classifier combination and meta-learning (stacked classifiers) to the problem of bootstrapping high accuracy automatic annotation of corpora with pronunciation information. The task we address in this paper consists of genera
ting phonemic representations reflecting the Flemish and Dutch pronunciations of a word on the basis of its orthographic representation (which in turn is based on the actual speech recordings). We compare several possible approaches to achieve the text-to-pronunciation mapping task: memory-based learning, transformation-based learning, rule induction, maximum entropy modeling, combination of classifiers in stacked learning, and stacking of meta-learners. We are interested both in optimal accuracy and in obtaining insight into the linguistic regularities involved. As far as accuracy is concerned, an already high accuracy level (93% for Celex and 86% for Fonilex at word level) for single classifiers is boosted significantly with additional error reductions of 31% and 38% respectively using combination of classifiers, and a further 5% using combination of meta-learners, bringing overall word level accuracy to 96% for the Dutch variant and 92% for the Flemish variant. We also show that the application of machine learning methods indeed leads to increased insight into the linguistic regularities determining the variation between the two pronunciation variants studied.
Bilingual terminologies are important resources for natural language processing (NLP) applications. The acquisition of bilingual terminology pairs is either human translation or automatic extraction from parallel data. We notice that comparable corpo
ra could also be a good resource for extracting bilingual terminology pairs, especially for e-commerce domain. The parallel corpora are particularly scarce in e-commerce settings, but the non-parallel corpora in different languages from the same domain are easily available. In this paper, we propose a novel framework of extracting bilingual terminologies from non-parallel comparable corpus in e-commerce. Benefiting from cross-lingual pre-training in e-commerce, our framework can extract the corresponding target terminology by fully utilizing the deep semantic relationship between source-side terminology and target-side sentence. Experimental results on various language pairs show that our approaches achieve significantly better performance than various strong baselines.
Legacy procedures for topic modelling have generally suffered problems of overfitting and a weakness towards reconstructing sparse topic structures. With motivation from a consumer-generated corpora, this paper proposes semiparametric topic model, a
two-step approach utilizing nonnegative matrix factorization and semiparametric regression in topic modeling. The model enables the reconstruction of sparse topic structures in the corpus and provides a generative model for predicting topics in new documents entering the corpus. Assuming the presence of auxiliary information related to the topics, this approach exhibits better performance in discovering underlying topic structures in cases where the corpora are small and limited in vocabulary. In an actual consumer feedback corpus, the model also demonstrably provides interpretable and useful topic definitions comparable with those produced by other methods.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
