Coding world model: rubric libraries

EdwardoSunny 's Collections

EM organisms

xArm tactile: direction + magnitude ablation (1 vs 9 arrows)

updated 10 days ago

Rubric libraries for a coding world model, by curation method; see https://codingwm.github.io/rubric_distribution.html

Upvote

EdwardoSunny/mle-rubrics-quota-final-pool

Viewer • Updated 10 days ago • 952 • 42

Note Quota · 952 rubrics
EdwardoSunny/pb-rubrics-quota-final-pool

Viewer • Updated 10 days ago • 628 • 37

Note Quota · 628 rubrics
EdwardoSunny/mle-rubrics-filtered-final

Viewer • Updated 10 days ago • 220 • 41

Note Filtered · 220 rubrics
EdwardoSunny/mle-rubrics-filtered-final-split

Viewer • Updated 10 days ago • 220 • 36

Note Filtered · 220 rubrics
EdwardoSunny/mle-rubrics-clean-v3

Viewer • Updated 10 days ago • 3.55k • 43

Note Unmatched (2nd+3rd collections) · 3,550 rubrics
EdwardoSunny/mle-rubrics-oracle-pool-matched

Viewer • Updated 10 days ago • 187 • 42

Note Oracle, pool-matched (reference) · 187 rubrics
EdwardoSunny/mle-rubrics-oracle-mle-matched

Viewer • Updated 10 days ago • 186 • 40

Note Oracle, MLE-bench-matched (reference) · 186 rubrics
EdwardoSunny/mle-rubrics-quota-pool

Viewer • Updated 10 days ago • 671 • 44

Note Quota, first version · 671 rubrics
EdwardoSunny/mle-rubrics-onpolicy-1000

Viewer • Updated about 1 month ago • 1k • 42

Note Wrong-answer (the original) · 1,000 rubrics
EdwardoSunny/mle-rubrics-clean

Viewer • Updated 16 days ago • 4.29k • 41

Note Unmatched · 4,294 rubrics
EdwardoSunny/mle-rubrics-operational-1k

Viewer • Updated 18 days ago • 963 • 30

Note Crash-rubric · 963 rubrics
EdwardoSunny/cwm-rubric-curation-metadata

Updated 10 days ago • 58

Note failure taxonomy, targets, rubric labels, mining cost

Upvote