Calibration round 2: rubric v2, agent-diff judge input, per-task dimensions, judge-matrix run plan - #75

Calibration round 2: rubric v2, agent-diff judge input, per-task dimensions, judge-matrix run plan#75

szjanikowski merged 59 commits into

mainNoesisVision/nasde-toolkit:mainfrom

calibration/ddd-weather-discount-v2NoesisVision/nasde-toolkit:calibration/ddd-weather-discount-v2Copy head branch name to clipboard