Share your thoughts, 1 month free Claude Pro on usSee more

Simulation deck generation on GEOS v1 (val)

92.1TreeSim

X+M

Updated 1mo ago

Evaluation Results

Method	Links
X+M 2026.06		92.1	0.011
S+M 2026.06		91.9	0.009
Self-Evolve 2026.06		91.9	0.009
S+X 2026.06		91.7	0.007
S+X+M 2026.06		91.1	0.001
Vanilla 2026.06		91	-
Self-Evolve-prose 2026.06		89.7	0.013
R+X 2026.06		89.3	0.017
R+M 2026.06		88.5	0.025
R+S+X+M 2026.06		88.5	0.025
R+S 2026.06		85.7	0.053