List your company

Multi-turn agent interaction on ALFWorld ScienceWorld WebShop Average

71.58Success Rate (SR)ATOD
-1.178417.710836.655.4892Jun 14, 2026Jun 16, 2026Jun 18, 2026Jun 20, 2026Jun 22, 2026Jun 24, 2026Jun 26, 2026

Evaluation results

MethodLinks
ATODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0671.58——
ATODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0671.06——
ATODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0670.62——
GRPOStudent Model=None, Teacher Model=Qwen3-4B GRPO2026.0668.93——
GRPOStudent Model=Qwen3-4B, Teacher Model=None2026.0668.93——
GRPOStudent Model=None, Teacher Model=Qwen3-30B-A3B GRPO2026.0668.91——
SODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0668.28——
OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0668.13——
OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0667.73——
TCODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0666.91——
TCODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0666.88——
TCODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0666.05——
SODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0665.74——
SODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0665.59——
OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0664.91——
SDARStudent Model=Qwen3-4B, Teacher Model=None2026.0664.37——
SDARStudent Model=Qwen3-1.7B, Teacher Model=None2026.0641.94——
GRPOStudent Model=Qwen3-1.7B, Teacher Model=None2026.0640.3——
GRPOStudent Model=Qwen3-0.6B, Teacher Model=None2026.0633.17——
Zero-shotParams=30B-A3B2026.0627.7241.417.96
SDARStudent Model=Qwen3-0.6B, Teacher Model=None2026.0625.44——
Guided-OPDParams=4B2026.0625.3742.6418.26
TCODParams=4B2026.0623.1139.518.71
Vanilla OPDParams=4B2026.0621.1338.6618.78
Guided-OPDParams=1.7B2026.0619.9337.9418.73
AdaSwitchParams=4B2026.0619.2537.0819.4
VanillaStudent Model=Qwen3-4B, Teacher Model=None2026.0619.24——
TCODParams=1.7B2026.0617.534.2819.63
Zero-shotParams=4B2026.0617.1118.419.3
Vanilla OPDParams=1.7B2026.0616.1132.2720.2
VanillaStudent Model=Qwen3-1.7B, Teacher Model=None2026.0615.59——
AdaSwitchParams=1.7B2026.0615.5231.9120.59
Guided-OPDParams=0.6B2026.0615.0134.0419.96
AdaSwitchParams=0.6B2026.0613.0126.4921.43
TCODParams=0.6B2026.0612.6231.0420.94
Vanilla OPDParams=0.6B2026.0611.325.1621.68
VanillaStudent Model=Qwen3-0.6B, Teacher Model=None2026.066.66——
Zero-shotParams=1.7B2026.066.5517.622.37
Zero-shotParams=0.6B2026.061.627.7823.51