Mixed reasoning tasks on Mixed workload n = 450

74.7Exact-match Accuracy

Flat

Updated 1mo ago

Evaluation Results

Method	Links
Flat 2026.06		74.7
accept-only ablation 2026.06		73.6
Transactional leadership 2026.06		73.1
Situational leadership 2026.06		71.8
Theory-free control 2026.06		70.9
Transformational leadership 2026.06		70.7
directives-only ablation 2026.06		70.4
Situational leadership 2026.06		70.4
Transformational leadership 2026.06		70.4
Theory-free control 2026.06		70
Flat 2026.06		69.8
Theory-free control 2026.06		69.8
Situational leadership 2026.06		69.8
accept-only ablation 2026.06		69.3
Transactional leadership 2026.06		69.1
Flat 2026.06		68.4
directives-only ablation 2026.06		68.4
directives-only ablation 2026.06		68.2
Transformational leadership 2026.06		67.8
Transactional leadership 2026.06		67.6
accept-only ablation 2026.06		67.6