Tokenizers
SentencePiece
Introduced by Kudo et al. in SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text ProcessingSentencePiece is a subword tokenizer and detokenizer for natural language processing. It performs subword segmentation, supporting the byte-pair-encoding (BPE) algorithm and unigram language model, and then converts this text into an id sequence guarantee perfect reproducibility of the normalization and subword segmentation.
Source: SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Language Modelling | 95 | 9.20% |
| Question Answering | 58 | 5.61% |
| Sentence | 47 | 4.55% |
| Decoder | 44 | 4.26% |
| Text Generation | 38 | 3.68% |
| Retrieval | 31 | 3.00% |
| Translation | 31 | 3.00% |
| Machine Translation | 28 | 2.71% |
| Natural Language Understanding | 20 | 1.94% |
BPE
