Search Results for author:
Found 5 papers, 5 papers with code
Embedding structure matters: Comparing methods to adapt multilingual vocabularies to new languages
Pre-trained multilingual language models underpin a large portion of modern NLP tools outside of English.
Planting and Mitigating Memorized Content in Predictive-Text Language Models
Language models are widely deployed to provide automatic text completion services in user products.
Learning to translate by learning to communicate
We formulate and test a technique to use Emergent Communication (EC) with a pre-trained multilingual model to improve on modern Unsupervised NMT systems, especially for low-resource languages.
Multilingual unsupervised sequence segmentation transfers to extremely low-resource languages
Further, we show that this transfer can be achieved by training over a collection of low-resource languages that are typologically similar (but phylogenetically unrelated) to the target language.
A Masked Segmental Language Model for Unsupervised Natural Language Segmentation
Segmentation remains an important preprocessing step both in languages where "words" or other important syntactic/semantic units (like morphemes) are not clearly delineated by white space, as well as when dealing with continuous speech data, where there is often no meaningful pause between words.
