3 ms·
I've noted that LLMs tend to produce defensive code to a fault. Lots of unnecessary checks, e.g. check for null/None/undefined multiple times for same valie. Th
by jampekka 1y ago
I've noted that LLMs tend to produce defensive code to a fault. Lots of unnecessary checks, e.g. check for null/None/undefined multiple times for same valie. This can lead to really hard to read code, even for the LLM itself.
The RL objectives probably heavily penalize exceptions, but don't reward much for code readability or simplicity.
- MakeAJiraTicket 1y agoI have a function that compares letters to numbers for the Major System and it's like 40 lines of code and copilot starts trying to add "guard rails" for "future proofing" as if we're adding more numbers or letters in the future. It's so annoying.