Arrows of Time for Large Language Models

6 点作者 tianlong超过 1 年前

2 条评论

nyoncore超过 1 年前

Isn't it obvious that since LLM are trained to predict the next word they do better than to predict the previous one?

评论 #39230234 未加载

评论 #39230205 未加载

tianlong超过 1 年前

There is a link with entropy creation?