Conceptual transfer: Using local classifiers for transfer selection
A key challenge for Machine Translation is transfer selection, i.e. to find the right translation for a given word from a set of alternatives (1:n). This problem becomes the more important the larger the dictionary is, as the number of alternatives increases. The contribution presents a novel approach for transfer selection, called conceptual transfer, where selection is done using classifiers based on the conceptual context of a translation candidate on the source language side. Such classifiers are built automatically by parallel corpus analysis: Creating subcorpora for each translation of a 1:n package, and identifying correlating concepts in these subcorpora as features of the classifier. The resulting resource can easily be linked to transfer components of MT systems as it does not depend on internal analysis structures. Tests show that conceptual transfer outperforms the selection techniques currently used in operational MT systems.
Code (0)
등록된 구현이 없습니다.
Tasks
Machine TranslationTranslationSimilar Papers 제목 키워드 기반
Transferable Perturbations of Deep Feature Distributions
Almost all current adversarial attacks of CNN classifiers rely on information derived from the output layer of the network. This work presents a new adversarial attack based on the modeling and exploitation of class-wise…
Adversarial AttackEasy Transfer Learning By Exploiting Intra-domain Structures
Transfer learning aims at transferring knowledge from a well-labeled domain to a similar but different domain with limited or no labels. Unfortunately, existing learning-based methods often involve intensive model select…
Computational EfficiencyDomain AdaptationModel SelectionTransfer LearningMulti-Stage Transfer Learning with an Application to Selection Process
In multi-stage processes, decisions happen in an ordered sequence of stages. Many of them have the structure of dual funnel problem: as the sample size decreases from one stage to the other, the information increases. A …
Transfer LearningGeoThreat: Transferable Targeted Adversarial Attacks on Large Vision-Language Models for Remote Sensing Image Interpretation
Adversarial attacks against large vision-language models (LVLMs) serve as an effective means of assessing their robustness in cross-modal semantic understanding. Existing studies mainly focus on corrupting visual inputs …
Adversarial AttackObject based Scene Representations using Fisher Scores of Local Subspace Projections
Several works have shown that deep CNN classifiers can be easily transferred across datasets, e.g. the transfer of a CNN trained to recognize objects on ImageNET to an object detector on Pascal VOC. Less clear, however, …
Scene Classification