List your company

Pass Rate Evaluation on ALFWorld (Interactive Agent)

74.15Pass@1EARLYEXP
24.146837.128450.1163.0916Mar 17, 2026

Evaluation results

MethodLinks
EARLYEXPBackbone=Llama3.1-8B2026.0374.1595.71
GRPO-RLVRBackbone=Llama3.1-8B2026.0372.594.29
EARLYEXPBackbone=Qwen2.5-7B2026.0372.2392.86
LEAFEBackbone=Llama3.1-8B2026.0371.7996.43
ACEBackbone=Llama3.1-8B2026.0370.3292.5
GRPO-RLVRBackbone=Qwen2.5-7B2026.0369.4691.43
LEAFEBackbone=Qwen2.5-7B2026.0367.594.29
ACEBackbone=Qwen2.5-7B2026.0366.3489.63
BASE (NO FT)Backbone=Llama3.1-8B2026.0329.8274.29
BASE (NO FT)Backbone=Qwen2.5-7B2026.0326.0778.57