Data Augmentation by Backtranslation (DAB) ヽ( •_-)ᕗ

Cross-lingual Language Model (XLM) pretraining and Model-Agnostic Meta-Learning (MAML) for fast adaptation of deep networks

Bangla Text Augmentation

Low resource machine translation using Transformers and Iterative Back translation

Implementing 5 Different Approaches To Augmenting Data For Natural Language Processing Tasks.

Source code and data for paper ``Unsupervised Dual Paraphrasing for Two-stage Semantic Parsing" in ACL 2020.

A PyTorch implementation of a transformer network trained using back-translation

An attempt to make Back-Translation differentiable, using probability weighted embeddings for predicted translations in the nucleus of the predicted distribution over target language sentences.

This repo offers a Python script using NLPAug library & RTT to augment text datasets. It processes TXT files in "data/" folder, translating text and creating augmented versions. Augmented data enhances NLP tasks like chatbot training & text classification. Includes overview of techniques, applications & implementation.

Cultural Translation Tool

Low resource language machine translation(az,be,tr -> en).

Tool to detect translation errors using GPT and back translation.

Code associated with the "Enhanced Cognitive Distortions Detection and Classification through Data Augmentation Techniques" paper

Common approaches to text augmentation, from random text-editing perturbations, back translation, to model-based transformations.

BackTranslation Experiment

A text generation library to paraphrase image captions using back translations or transfer learning.

Noise Identification, Noise reduction, and Sentiment Analysis on Bangla Noisy Texts