Re:OpenAI O3-Mini
by m0rde on Saturday February 01, 2025 @11:29AM
(#42899575)
Would you elaborate on why counting braces is different from counting spaces to determine hierarchy? Or is it more about the repetition of higher levels keys in chunks (which could be done in JSON)?
[ Parent
| on HN ]
Repetition of topics and subtopics is by far the most important part, reinforcing attention on the topic at hand even if the text in the chunk appears unrelated to the topic when viewed in isolation. Keeping the indentation is mostly about convenience - not having to postprocess the YAML beyond repeating higher level keys.
Keeping the indentation is also important because it is an implicit and repeated indication of the nesting level of the content that follows (LLMs have trouble with balancing nested parentheses as the sibling comment to yours explains) and LLMs have been exposed to large amounts of text where indentation matters (such as Python code and lists of bullet points) during training, so they have learned to handle this well.