Well, o3-mini-high just successfully found the root cause of a seg fault that o1 missed: mistakenly using _mm512_store_si512 for an unaligned store that should have been _mm512_storeu_si512.
I think that "Good Forth programmers arrange things so that they flow on the stack" has analogs in other languages. For example, arranging things in J so that short tacit expressions naturally provide the functions you need.
shows Sonnet 3.5 using the Google web UI in an automated fashion. Do Google's terms really permit this? Will Google permit this when it is happening at scale?
These strategies seem immediately practical. If you want to go beyond zero-shot for LLM coding, you may not need a complicated agent architecture - just start with escalation, retry, and warming.
APPS has 3 subsets by difficulty level: introductory, interview, and competition. It isn't clear which subset Claude 3 was benchmarked on. Even if it is just "introductory" it is still pretty good, but it would be good to know.
Push has remained active for genetic programming, and there's recent work comparing program synthesis with language models vs. PushGP: https://arxiv.org/pdf/2304.10423.pdf
This looks promising. For me, it successfully scraped a webpage that Bing Chat would not. A couple comments: (a) preserve newlines in the output (I'm trying to get a CSV output), (b) the output length appears to be truncated - too short for my application.
The article expresses disappointment that a <40 character APL program mostly just demonstrates "why Game of Life is perfectly suited to solving in APL". But it seems harder to argue Game of Life is "perfectly suited" to solving with low-level x86 machine language.
Does anyone know of other places where the word "flow" is used like this?:
> Good Forth programmers arrange things so that they flow on the stack.
I think this concept of "flow" applies elsewhere, for example when creating tacit programs in J. But I haven't seen "flow" used like this anywhere except this article.