Hacker News Reader

I've heard that LLMs can perform worse with these more efficient representations compared to e.g. JSON, because they've seen far fewer examples of them during training. Do you know how true that is?

View on HN

Absolutely, but usually when working with a bespoke format for optimization, it's paired with an LLM specifically trained on that format.

You are right about cryptic formats. CommerceTXT is semantically structured. Models like GPT, Claude and Gemini understand it out-of-the-box via ICL.

Reply on news.ycombinator.com