Share on Facebook Tweet on Twitter Share on LinkedIn Share by email
> Publications > Less is more: significance-based N-gram selection for smaller, better language models