FrontierThe story, in brief
Better language models and their implications
STATE-OF-THE-ART. OpenAI just released a large-scale language model that nails benchmarks and handles reading comprehension, translation, and Q&A without task-specific training.
The KeyNews take
Why it matters
This demonstrates a fundamental shift in AI capabilities: a single unsupervised model achieving broad competence across multiple language tasks, signaling the viability of general-purpose language models as a platform for downstream applications.
The key facts
5 to knowLarge-scale unsupervised language model trained
State-of-the-art performance on language modeling benchmarks
Rudimentary capabilities across reading comprehension, machine translation, question answering, and summarization
No task-specific training required
Published February 14, 2019
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: We’ve trained a large-scale unsupervised language model which generates coherent paragraphs of text, achieves state-of-the-art performance on many language modeling benchmarks, and performs rudimentary reading comprehension, machine translation, question answering, and summarization—all without…