A reproduction of training T5 on cLang-8 (corresponding to Table 4) in the following paper: A Simple Recipe for Multilingual Grammatical Error Correction.

The code and the performance on GEC benchmarks are avaliable from https://github.com/gotutiyan/gec-t5.

The same as cLang-8 corpus and the original Lang-8 corpus, the pre-trained models are distributed for research and educational purposes only.

Downloads last month
23
Safetensors
Model size
77M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Collection including gotutiyan/gec-t5-small-clang8