CoEdit-xl Open-Source Text Editing Model - Freely Achieve English Text Grammar Correction and Style Transformation

Coedit Xl

Developed by grammarly

CoEdIT-xl is a text editing model fine-tuned on the CoEdIT dataset based on the google/flan-t5-xl model, supporting various editing tasks for English text such as grammar correction and style transformation.

Large Language Model

Transformers

English#Text Editing Optimization #Instruction Fine-tuning #Grammar Correction

Downloads 374

Release Time : 5/11/2023

Model Overview

This model is obtained by fine-tuning the flan-t5-xl model on the CoEdIT dataset, specifically designed to perform various text editing tasks such as grammar correction and style transformation.

Model Features

Multi-task Text Editing

Supports various text editing tasks, including grammar correction, style transformation, simplification, etc.

Instruction Fine-tuning

Fine-tuned with task-specific instructions, capable of understanding and executing complex editing commands.

Large-scale Pre-training

Based on the 3-billion-parameter flan-t5-xl model, it has strong language understanding and generation capabilities.

Model Capabilities

Grammar error correction

Text style transformation

Text simplification

Instruction-based text editing

Text quality improvement

Use Cases

Writing Assistance

Grammar Correction

Automatically detects and corrects grammatical errors in text

Improves the grammatical accuracy of the text

Style Adjustment

Adjusts the formality or style of the text according to instructions

Makes the text more suitable for the target audience or scenario

Content Optimization

Text Simplification

Simplifies complex text into a more understandable version

Improves the readability of the text

🚀 Model Card for CoEdIT-xl

This model is obtained by fine - tuning the google/flan - t5 - xl model on the CoEdIT dataset, enabling efficient text editing.

🚀 Quick Start

This model was obtained by fine - tuning the corresponding google/flan - t5 - xl model on the CoEdIT dataset. Details of the dataset can be found in our paper and repository.

Paper: CoEdIT: Text Editing by Task - Specific Instruction Tuning

Authors: Vipul Raheja, Dhruv Kumar, Ryan Koo, Dongyeop Kang

✨ Features

Model Details

Model Description

Property	Details
Language(s) (NLP)	English
Finetuned from model	`google/flan - t5 - xl`

Model Sources

Property	Details
Repository	https://github.com/vipulraheja/coedit
Paper	https://arxiv.org/abs/2305.09857

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

We make available the models presented in our paper.

Model	Number of parameters
CoEdIT - large	770M
CoEdIT - xl	3B
CoEdIT - xxl	11B

Advanced Usage

Text Revision Task

Given an edit instruction and an original text, our model can generate the edited version of the text.

task_specs

from transformers import AutoTokenizer, T5ForConditionalGeneration

tokenizer = AutoTokenizer.from_pretrained("grammarly/coedit-xl")
model = T5ForConditionalGeneration.from_pretrained("grammarly/coedit-xl")
input_text = 'Fix grammatical errors in this sentence: When I grow up, I start to understand what he said is quite right.'
input_ids = tokenizer(input_text, return_tensors="pt").input_ids
outputs = model.generate(input_ids, max_length=256)
edited_text = tokenizer.decode(outputs[0], skip_special_tokens=True)

Software

https://github.com/vipulraheja/coedit

📚 Documentation

Uses

The model is mainly used for the text revision task, where it can generate an edited version of the text based on the given edit instruction and original text.

📄 License

This model is released under the CC - BY - NC - 4.0 license.

📚 Additional Information

Datasets

grammarly/coedit
facebook/asset
jfleg
zaemyung/IteraTeR_plus
wi_locness
GEM/wiki_auto_asset_turk
discofuse

Metrics

sari
bleu
accuracy

📚 Citation

BibTeX:

@article{raheja2023coedit,
      title={CoEdIT: Text Editing by Task-Specific Instruction Tuning}, 
      author={Vipul Raheja and Dhruv Kumar and Ryan Koo and Dongyeop Kang},
      year={2023},
      eprint={2305.09857},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

APA: Raheja, V., Kumar, D., Koo, R., & Kang, D. (2023). CoEdIT: Text Editing by Task - Specific Instruction Tuning. ArXiv. /abs/2305.09857

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご