Stephen with a ph...D
@stephenmcgeephd.bsky.social
Data nerd focused on integrating AI in into healthcare genetics
Shoutout to DeepSeek R1 for surpassing OpenAI’s o1 model in performance - and for shaking up the AI world, in general.
February 2, 2025 at 2:28 PM
Shoutout to DeepSeek R1 for surpassing OpenAI’s o1 model in performance - and for shaking up the AI world, in general.
Despite the dramatic title, they do admit, “HLE tests structured academic problems rather than open-ended research or creative problem-solving abilities…HLE may be the last academic exam we need to give to models, but it is far from the last benchmark for AI.”
February 2, 2025 at 2:28 PM
Despite the dramatic title, they do admit, “HLE tests structured academic problems rather than open-ended research or creative problem-solving abilities…HLE may be the last academic exam we need to give to models, but it is far from the last benchmark for AI.”