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

Action-component payoff optimization on L3 warmup (off-diagonal)

4.17Per-Interaction Payoff

Reciprocity Gradient

0.09321.15162.213.2684May 8, 2026
Updated 22d ago

Evaluation Results

MethodLinks
4.1793
4.0891
2026.05
1.6336
2026.05
1.5835
2026.05
1.4833
2026.05
0.5512
2026.05
0.327
2026.05
0.256