Hacker News Reader

Comment by throwatdem12311 | Hacker News Reader

I hate the way Claude writes so much I’m at a breaking point ready ti crash out and try to convince my boss to switch to literally anything else.

It drives me up the fucking wall.

The worst part is I can prompt Claude with “this is gobbledegook, simplify” and it will reword its previous answer perfectly but no amount of hooks or system prompt hacking will fix it otherwise.

View on HN

Try using a subagent loop where you have one agent generate text and another critique the text according to some guidelines/goal.

The tech bro speak gets so insane that, if I paste it to another claude session, it's so unintelligible that new claude session doesn't even know what the other was trying to say.

It'll be interesting to see if all this causes a subtle shift in vernacular.

Copy/pasting Claude output to itself in another session is one of the easiest ways to demonstrate model collapse.

It’s also why I’m not a fan of using subagents for code review like Claude Code started doing by default recently. It’s getting output from another LLM, then summarizing it for me so by the time I see it it’s already been blended up once.

Can you explain what “one of the easiest ways to demonstrate model collapse” means, please?

I think you mean “if it can’t understand its own output this demonstrates model collapse” but I’m not entirely sure what “model collapse” means in this context.

Reply on news.ycombinator.com