A negative result, its root cause, and the feature it produced. My last post introduced ctxfold —...
A negative result, its root cause, and the feature it produced.
My last post introduced ctxfold — lossless, structure-aware compression for the bulky stuff we put in LLM prompts. Its benchmark table had one cell I wasn't proud of:
| CSV / TSV | char reduction | ~30–45%* |
*CSV readability not yet validated against a model
Enter fullscreen mode
Most prompt compression discussions focus on the happy path: you have a long RAG context, you trim it...
Disclosure up front: I maintain one of the five libraries tested (SignalTree), and it's the one that...
I logged 12,400 structured-output calls. 1,742 came back wrapped in markdown fences despite a strict system prompt. So I wrote a…
I've been building data2prompt for a few months now. It takes a data-heavy project (CSVs, SQL dumps,...
Cut LLM prompt tokens on structured data — losslessly A small, dependency-free tool for...
A 1.5B model running on your laptop will return JSON that almost parses. The closing brace is...
