List your company
ResearchFormerly WizWand

Multi-agent coordination on Navigation ablation Clean (test)

2.43Task Metric ValueVL
1.43161.69081.952.2092May 13, 2026
Updated 5mo ago

Evaluation results

MethodLinks
VLAttack rate=0.252026.052.43
MLAttack rate=0.752026.052.41
STAttack rate=0.252026.052.41
J-MAttack rate=0.252026.052.31
J-WAttack rate=0.252026.052.3
CBTSAttack rate=0.252026.052.16
MMRAttack rate=0.752026.052.09
MMRAttack rate=0.252026.052.06
VLAttack rate=0.52026.052.02
MLAttack rate=0.52026.052
MMRAttack rate=0.52026.051.98
STAttack rate=0.52026.051.95
NSAttack rate=0.52026.051.88
CBTSAttack rate=0.52026.051.85
NSAttack rate=0.252026.051.8
MLAttack rate=0.252026.051.79
NSAttack rate=0.752026.051.76
J-WAttack rate=0.52026.051.73
J-MAttack rate=0.52026.051.6
VLAttack rate=0.752026.051.56
CBTSAttack rate=0.752026.051.55
J-WAttack rate=0.752026.051.53
J-MAttack rate=0.752026.051.51
STAttack rate=0.752026.051.47