Don't Force Your LLM to Write Terse [Q/Kdb] Code: An Information Theory Argument

Re:Don't Force Your LLM to Write Terse [Q/Kdb] Code: An Info... by segmondy on Monday October 20, 2025 @09:55AM (#45643985)

First pass on my local deepseekv3.1-Terminus at Q4 answered it correctly. if anything, i think LLMs should write terse code, Q/J/APL/Forth/Prolog/Lisp, tokens is precious. It's insane to waste precious tokens generating Java, javascript and other overly verbose code...

https://pastebin.com/VVT74Rp9

[ Parent | on HN ]

Re:Don't Force Your LLM to Write Terse [Q/Kdb] Code: An Info... by gabiteodoru on Monday October 20, 2025 @01:50PM (#45646897)

It did go back on itself 3 times, no? "Actually, let’s trace for x=3:" (it had just computed for x=3 the first time); then "Better to check actual q output:" -- did it actually run it in a q session, or just pretended? And another one "That doesn’t seem to align. Let’s do it step by step:"

[ Parent | on HN ]