: A large-scale dataset containing approximately 92,000 computer science papers from 31 major conferences. It includes AI-generated summaries (GPT-3.5) designed for large-scale scientometric studies and automated literature reviews.
: A dataset for stuttering event detection containing 28k labeled clips from podcasts. It is often used to train models to identify blocks, prolongations, and repetitions in speech. You can find it on GitHub via Apple's ML research . Download 273k txt
: A massive collection of 1.14 billion content regions from historical American newspaper articles. It is used for training large language models (LLMs) and exploring world history. It is often used to train models to
If you are looking for "txt" files related to AI crawling, you might be interested in the proposal. It is used for training large language models
However, based on your interest in downloading text datasets and finding helpful papers, here are several prominent datasets of similar scale or naming conventions that are frequently used in research: Related Research Datasets
: A proposed standard Markdown file placed at a website's root to serve as a curated, distraction-free index for Large Language Models to crawl.