Nanolang: A tiny experimental language designed to be targeted by coding LLMs
1–10 of 210 posts
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#2So, then if I want to use a certain terminal text editor to create a clone of it in nanolang, I'd end up typing nano nano.nano on the command line.
I might accidentally summon a certain person from Ork.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#3One novel part here is every function is required to have tests that run at compile time.
I'm still skeptical of the value add having to teaching a custom language to an LLM instead of using something like lua or python and applying constraints like test requirements onto that.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#4Looks nice!
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#5It seems that something that does away with human friendly syntax and leans more towards a pure AST representation would be even better? Basically a Lisp but with very strict typing might do the trick. And most LLMs are probably trained on lots of Lisps already.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#6Developed by Jordan Hubbard of NVIDIA (and FreeBSD).
My understanding/experience is that LLM performance in a language scales with how well the language is represented in the training data.
From that assumption, we might expect LLMs to actually do better with an existing language for which more training code is available, even if that language is more complex and seems like it should be “harder” to understand.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#7Developed by Jordan Hubbard of NVIDIA (and FreeBSD). My understanding/experience is that LLM performance in a language scales with how well the language is represented in the training data. From that assumption, we might expect LLMs to actually do better with an existing language for which more training code is available, even if that language is more complex and seems like it should be “harder” to understand.
easy enough to solve with RL probably
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#8Developed by Jordan Hubbard of NVIDIA (and FreeBSD). My understanding/experience is that LLM performance in a language scales with how well the language is represented in the training data. From that assumption, we might expect LLMs to actually do better with an existing language for which more training code is available, even if that language is more complex and seems like it should be “harder” to understand.
I think it's depressingly true of any novel language/framework at this point, especially if they have novel ideas.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#9I feel like the time for this was two years ago, and LLMs are now less bothered by remembering syntax than I am. It's a nice lisp-y syntax though.
Re: Nanolang: A tiny experimental language designed to be targeted by coding LLMs
#10Developed by Jordan Hubbard of NVIDIA (and FreeBSD). My understanding/experience is that LLM performance in a language scales with how well the language is represented in the training data. From that assumption, we might expect LLMs to actually do better with an existing language for which more training code is available, even if that language is more complex and seems like it should be “harder” to understand.
A lot of this depends on your workflow. A language with great typing, type checking and good compiler errors will work better in a loop than one with a large surface overhead and syntax complexity, even if it's well represented. This is the instinct behind, e.g. https://github.com/toon-format/toon, a json alternative format. They test LLM accuracy with the format against JSON, (and are generally slightly ahead of JSON).
Additionally just the ability to put an entire language into context for an LLM - a single document explaining everything - is also likely to close the gap.
I was skimming some nano files and while I can't say I loved how it looked, it did look extremely clear. Likely a benefit.