8 papers with code • 4 benchmarks • 2 datasets
Using data and models available for one language for which ample such resources are available (e.g., English) to solve a natural language inference task in another, commonly more low-resource, language.
We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers.
Ranked #1 on Question Answering on CoQA
Although widely adopted, existing approaches for fine-tuning pre-trained language models have been shown to be unstable across hyper-parameter settings, motivating recent work on trust region methods.
We introduce an architecture to learn joint multilingual sentence representations for 93 languages, belonging to more than 30 different families and written in 28 different scripts.
State-of-the-art natural language processing systems rely on supervision in the form of annotated data to learn competent models.
Ranked #3 on Natural Language Inference on XNLI French
Many modern NLP systems rely on word embeddings, previously trained in an unsupervised manner on large corpora, as base features.
Ranked #1 on Semantic Textual Similarity on SentEval
In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences.
Ranked #1 on Question Answering on TweetQA (ROUGE-L metric)
This dataset, named FarsTail, includes 10, 367 samples which are provided in both the Persian language as well as the indexed format to be useful for non-Persian researchers.
Ranked #1 on Natural Language Inference on FarsTail
While monolingual word embeddings encode information about words in the context of a particular language, cross-lingual embeddings define a multilingual space where word embeddings from two or more languages are integrated together.