List your company

Interactive LLM Agent Task Success on ALFWorld Seen

89.6Success RateQ-Evolve
76.679.97583.3586.725Jun 5, 2026

Evaluation results

MethodLinks
Q-EvolveBackbone=Llama-3-8B-Instruct2026.0689.6
ETO + MPOBackbone=Llama-3-8B-Instruct2026.0685
SFT + MPOBackbone=Llama-3-8B-Instruct2026.0680.7
KnowAgentBackbone=Llama-3-8B-Instruct2026.0680
SFTBackbone=Llama-3-8B-Instruct2026.0679.3
WKMBackbone=Llama-3-8B-Instruct2026.0677.5
ETOBackbone=Llama-3-8B-Instruct2026.0677.1