For those who are knowledgeable about the field but not yet the author of this post, it is worth mentioning that Shunyu Yao has played a huge role in the development of LLM-based AI agents, including being an author / contributor to:
My intention is to send free digital updates to those who purchased the original 2012 PDF via the website, I have a list via fetchapp.com. I will probably remove this message for the new version. Print editions or other proprietary ebook formats (like kindle, etc) will be for purchase only for the new version. More: https://github.com/nature-of-code/natureofcode.com/pull/42
I think what they mean is that if your post gets to the front-page via the second-chance pool and then gets 100 upvotes, your karma increases only by 50.
Wow, I wasn't expecting to find a reference to a horror manga in the comments of my post about (what I believed to be) innocuous spirals! That's amazing, although I'm too squeamish to try and read it!
I found this very interesting too, and it actually has a great explanation! Maybe I should have even explored this side a bit deeper in the post itself.
Let's say the spiral has rotated 6000 degrees, and I'm approximating it with 100 points (one point every 60 degrees) and line segments. Well, a hexagon is nothing other than 6 points chosen 60 degrees apart from each other at the same distance from a central point, connected with straight lines. The same thing holds for a square at, e.g. 9000 degrees.
Thanks a lot! I agree that there is great educational potential in this format. I've definitely learned some cool things myself from the posts at https://explorabl.es/
- ReAct
- Reflexion
- SWE-bench
- OpenAI Deep Research
- OpenAI Operator