← all repositories

patil-suraj/question_generation

A project that fine-tunes transformer models like T5 for neural question generation from text passages.

1.1k stars Jupyter Notebook Language Models
question_generation
Velocity · 7d
+0.5
★ / day
Trend
steady
star history

This repository provides open-source implementations for neural question generation using pre-trained seq2seq transformers. It supports multiple approaches including answer-aware QG, multitask QA-QG, and end-to-end answer-agnostic generation. The project offers fine-tuning scripts, evaluation pipelines, and pretrained checkpoints for the T5 model family applied to question generation.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.