highsierralabs / Laconic_testsView on GitHub
A small A/B test of a laconic brevity directive on three LLMs: 54 replies to safety decisions, blinded two-scorer consensus against pre-registered keys. Verdicts survived compression; second-tier checks often didn't. n=3 per cell, single scorer, one domain - a worked protocol, not a benchmark.
☆17Aug 10, 2026Updated 2 months ago

Alternatives and similar repositories for Laconic_tests

Users that are interested in Laconic_tests are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?