jpelhaw/t5-word-sense-disambiguation

The T5-large for Word Sense Disambiguation has been trained on the SemCor 3.0 dataset . The model can be loaded to perform a few-shot classification like so: import AutoModelForSeq2SeqLM, AutoTokenizerAutoModel forSeq 2.0 and tokenizer.tokenize(input, add_special_tokens=True) The model is the checkpoint for T4-large after being trained on a SemCor dataset . It can be used to identify a word in the following context: " java " or " an island in Indonesia to the south of Borneo " in the . south of the . north of B,