Such claims are still more hype than reality! The actual paper itself is a hypothetical (see link at the bottom).
Even today's best large language models can not really do the job! What about the usual accompanying tables and charts just to give one example!
Subject experts would also still be able to discern the novelty of any parts of the content of the paper.
We Asked GPT-3 to Write an Academic Paper about Itself—Then We Tried to Get It Published An artificially intelligent first author presents many ethical questions—and could upend the publishing process
No comments:
Post a Comment