Fichier de travail (INPUT) : ./DUMP-TEXT/2-21.txt
Encodage utilisé (INPUT) : utf-8
Forme recherchée : translation|traduction|机器
_________________________________________________________________________________________________
Ligne n°1 : ... #VentureBeat Feed VentureBeat » IBM, Harvard develop tool to tackle- Ligne n°2 : black box problem in AI translation Comments Feed alternate alternate
- Ligne n°104 : IBM, Harvard develop tool to tackle black box problem in AI translation
- Ligne n°125 : In recent years, machine translation has improved immensely thanks to
Ligne n°126 : advances in deep learning and neural networks. However, the advantages ...
Ligne n°145 : ... Seq2Seq-Vis is focused on “sequence-to-sequence” models, the AI- Ligne n°146 : architecture used in most modern machine translation systems.
Ligne n°147 : “Sequence-to-sequence models can learn to transform an arbitrary-length ...
Ligne n°149 : ... Strobelt, scientist at IBM Research, adding that aside from language- Ligne n°150 : translation, sequence-to-sequence is also used in other fields such as
Ligne n°151 : question-answering, summarization of long text and image captioning. ...- Ligne n°155 : In a nutshell, sequence-to-sequence translation models run a source
Ligne n°156 : string through several neural networks to map it to the target language ...- Ligne n°161 : Visualizing machine translation
Ligne n°172 : ... Stobelt showed us how the tool works on its demo website, which has an- Ligne n°173 : example of a German-to-English translation gone wrong. The sentence
Ligne n°174 : “die längsten reisen fangen an , wenn es auf den straßen dunkel wird .” ...
Ligne n°179 : ... Seq2Seq-vis creates a visual representation of the different stages of- Ligne n°180 : the sequence-to-sequence translation process. This enables the user to
Ligne n°181 : examine the model’s decision process and find where the error is taking ...
Ligne n°198 : ... in the “beam search,” the AI model that refines the output of the- Ligne n°199 : translation model.
- Ligne n°226 : But using Seq2Seq-Vis is not for the end user of translation
Ligne n°227 : applications. It requires general knowledge of how sequence-to-sequence ...