List your company

Multi-turn information acquisition on Home

89.78Success Rate (SR)CUP
15.981635.140854.373.4592Apr 5, 2026

Evaluation results

MethodLinks
CUPModel=Llama-3.1-8B2026.0489.783.75
CUPModel=Qwen3-4B2026.0484.143.67
CUPModel=Mistral-7B-v0.32026.0483.873.62
MISQ-HFModel=Llama-3.1-8B2026.04754.12
MISQ-HFModel=Mistral-7B-v0.32026.0471.774.06
MISQ-HFModel=Qwen3-4B2026.0469.894.1
BED-LLMModel=Llama-3.1-8B2026.0469.354.35
BED-LLMModel=Qwen3-4B2026.0465.054.31
BED-LLMModel=Mistral-7B-v0.32026.0463.444.34
UoTModel=Llama-3.1-8B2026.0460.484.61
UoTModel=Qwen3-4B2026.0459.684.69
UoTModel=Mistral-7B-v0.32026.0459.134.71
ATDModel=Llama-3.1-8B2026.0456.994.56
ATDModel=Mistral-7B-v0.32026.0454.34.5
ATDModel=Qwen3-4B2026.0453.764.58
LLM-planningModel=Mistral-7B-v0.32026.0452.694.62
LLM-planningModel=Qwen3-4B2026.0451.344.65
LLM-planningModel=Llama-3.1-8B2026.0449.194.7
CoTModel=Qwen3-4B2026.0448.393.88
CoTModel=Llama-3.1-8B2026.0447.313.94
CoTModel=Mistral-7B-v0.32026.0445.983.93
Direct PromptingModel=Qwen3-4B2026.0441.924.05
Direct PromptingModel=Mistral-7B-v0.32026.0439.524.11
Direct PromptingModel=Llama-3.1-8B2026.0438.174.15
SBERT2026.0418.824.73