List your company

Logical Reasoning on RP LP Domain, Npred ≤ 30, δ ≤ 6

100Accuracydirect w. L = 128
83.3687.689296.32May 5, 2026

Evaluation results

MethodLinks
direct w. L = 128Reasoning Strategy=Direct, Number of Layers=1282026.05100
cot w. L = 64Reasoning Strategy=CoT, Number of Layers=642026.05100
direct w. L = 24Reasoning Strategy=Direct, Number of Layers=242026.05100
direct w. L = 128Reasoning Strategy=Direct, Number of Layers=1282026.05100
cot w. L = 64Reasoning Strategy=CoT, Number of Layers=642026.05100
cot w. L = 128Reasoning Strategy=CoT, Number of Layers=1282026.05100
direct w. L = 48Reasoning Strategy=Direct, Number of Layers=482026.0599.9
direct w. L = 64Reasoning Strategy=Direct, Number of Layers=642026.0599.9
cot w. L = 48Reasoning Strategy=CoT, Number of Layers=482026.0599.9
cot w. L = 96Reasoning Strategy=CoT, Number of Layers=962026.0599.9
cot w. L = 128Reasoning Strategy=CoT, Number of Layers=1282026.0599.9
direct w. L = 96Reasoning Strategy=Direct, Number of Layers=962026.0599.8
cot w. L = 32Reasoning Strategy=CoT, Number of Layers=322026.0599.8
cot (baseline)Reasoning Strategy=CoT2026.0599.7
direct w. L = 32Reasoning Strategy=Direct, Number of Layers=322026.0599.6
cot w. L = 24Reasoning Strategy=CoT, Number of Layers=242026.0599.6
direct w. L = 24Reasoning Strategy=Direct, Number of Layers=242026.0599.3
cot w. L = 16Reasoning Strategy=CoT, Number of Layers=162026.0598.7
cot (baseline)Reasoning Strategy=CoT2026.0597.4
direct (baseline)Reasoning Strategy=Direct2026.0596.9
direct w. L = 16Reasoning Strategy=Direct, Number of Layers=162026.0596.6
cot w. Nepochs = 60Reasoning Strategy=CoT, N_epochs=602026.0596.5
cot w. Nepochs = 30Reasoning Strategy=CoT, N_epochs=302026.0596
direct w. Nepochs = 60Reasoning Strategy=Direct, N_epochs=602026.0587.2
direct (baseline)Reasoning Strategy=Direct2026.0585.7
direct w. Nepochs = 30Reasoning Strategy=Direct, N_epochs=302026.0584