AI Technology & Industry Review
A research team from the University of Melbourne, Facebook AI, and Twitter Cortex proposes a black-box test method for assessing and debugging the numerical translation of neural machine translation systems in a systematic manner. The approach reveals novel types of errors that are general across multiple state-of-the-art translation systems.
A research team from ByteDance AI Lab, University of Wisconsin–Madison and Nanjing University wins the ACL 2021 best paper award. Their proposed Vocabulary Learning via Optimal Transport (VOLT) approach leverages optimal transport to automatically find an optimal vocabulary without trial training.
Microsoft Research Asia (MSRA) has been dubbed the “Whampoa Academy for AI” in reference the elite Chinese military school. MSRA is a bootcamp for NLP research and has trained more than 500 interns, 20 PhDs and 20 postdocs over the past two decades.
Tsinghua Natural Language Processing Group (THUNLP) has published a great reading list for any budding AI researchers whose New Year’s resolution is to study machine translation. The list compiles the most influential machine translation papers from the past 30 years, spotlighting the 10 most important contributions to the development of machine translation.
The author uses a hard-attention (Viterbi alignment matrix) of the SMT (Statistical Machine Translation) to help learn a better alignment matrix in NMT during the training.
In this paper, the author aims to design a smart way to handle rare/unknown words in the bilingual training data during machine translation.
This paper is the implementation of ‘encoder-decoder reconstructor framework’ for neural machine translation for the English-Japanese translation task.
Compared to SMT, NMT can train multiple features jointly and does not need prior domain knowledge, enabling zero-shot translation. In addition to higher BLEU score and better sentence structure, NMT can also help reduce morphology errors, syntax errors, and word order errors of SMT.
Although no innovations for the NMT architecture was introduced, the authors claim that a classic NMT baseline system with carefully tuned hyperparameters can still achieve comparable result to the state-of-the-art.
Instead of exploring target-side monolingual data for NMT, the author of this paper proposes two novel methods for source-side monolingual data exploration.
This talk was recently given by Prof. Christopher Manning at Simons Institute, UC Berkeley. It is an introductory tutorial without complicated algorithms.