List your company

Multi-turn agent interaction on ALFWorld IID

43.57Success Rate (SR)Zero-shot
-1.493210.205921.90533.6041Jun 14, 2026

Evaluation results

MethodLinks
Zero-shotParams=30B-A3B2026.0643.5722.46
Guided-OPDParams=4B2026.0637.1422.76
Guided-OPDParams=1.7B2026.0632.8623.34
TCODParams=4B2026.0632.8623.18
Vanilla OPDParams=4B2026.0631.4323.36
AdaSwitchParams=4B2026.0629.2923.59
Zero-shotParams=4B2026.0628.5724.25
TCODParams=1.7B2026.0627.8623.87
Guided-OPDParams=0.6B2026.0627.1424.5
AdaSwitchParams=1.7B2026.0625.7124.83
AdaSwitchParams=0.6B2026.0625.1124.93
TCODParams=0.6B2026.0624.2925.32
Vanilla OPDParams=1.7B2026.0623.5725.45
Vanilla OPDParams=0.6B2026.0619.2926.24
Zero-shotParams=1.7B2026.068.128.35
Zero-shotParams=0.6B2026.060.2430