List your company

Reconstruction Error Analysis on Agent-eval 8 templates user agent GPT-5.4 nano (aggregate)

5.6wMAPEM-Opt
1.22430.76260.389.838May 4, 2026

Evaluation results

MethodLinks
M-Optε=0.1, B=3k + 12026.055.6
M-Optε=0.1, B=2k + 12026.057.6
M-Rootsε=0.1, B=3k + 12026.057.9
M-Optε=0.05, B=3k + 12026.0510.5
M-Rootsε=0.1, B=2k + 12026.0510.6
M-Optε=0.1, B=k + 12026.0512.3
M-Rootsε=0.05, B=3k + 12026.0513.9
M-Optε=0.05, B=2k + 12026.0514
M-Rootsε=0.1, B=k + 12026.0515.9
M-Rootsε=0.05, B=2k + 12026.0517.8
M-Optε=0.05, B=k + 12026.0520.9
M-Allε=0.1, B=k + 12026.0521.6
M-Allε=0.1, B=3k + 12026.0521.7
M-Allε=0.1, B=2k + 12026.0522.2
M-Rootsε=0.05, B=k + 12026.0526
M-Allε=0.05, B=k + 12026.0532.8
M-Allε=0.05, B=2k + 12026.0534.4
M-Allε=0.05, B=3k + 12026.0534.6
M-Optε=0.01, B=3k + 12026.0535.5
M-Rootsε=0.01, B=3k + 12026.0543.2
M-Optε=0.01, B=2k + 12026.0547.2
M-Rootsε=0.01, B=2k + 12026.0555.5
M-Optε=0.01, B=k + 12026.0571.6
M-Rootsε=0.01, B=k + 12026.0582.6
M-Allε=0.01, B=k + 12026.0598.8
M-Allε=0.01, B=2k + 12026.05108
M-Allε=0.01, B=3k + 12026.05115