List your company

Multi-turn information acquisition on Inspired

92.86Success Rate (SR)CUP
7.954429.997252.0474.0828Apr 5, 2026

Evaluation results

MethodLinks
CUPModel=Llama-3.1-8B2026.0492.863.86
CUPModel=Mistral-7B-v0.32026.0491.844.01
CUPModel=Qwen3-4B2026.0489.83.93
MISQ-HFModel=Llama-3.1-8B2026.0477.554.32
MISQ-HFModel=Mistral-7B-v0.32026.0474.494.28
BED-LLMModel=Llama-3.1-8B2026.0471.434.39
MISQ-HFModel=Qwen3-4B2026.0467.354.3
BED-LLMModel=Mistral-7B-v0.32026.0466.334.58
BED-LLMModel=Qwen3-4B2026.0465.314.41
UoTModel=Llama-3.1-8B2026.0462.234.71
UoTModel=Mistral-7B-v0.32026.0461.224.66
UoTModel=Qwen3-4B2026.0459.184.7
ATDModel=Qwen3-4B2026.0458.164.68
ATDModel=Llama-3.1-8B2026.0458.164.64
ATDModel=Mistral-7B-v0.32026.0457.144.63
LLM-planningModel=Llama-3.1-8B2026.04504.77
LLM-planningModel=Qwen3-4B2026.0448.984.73
LLM-planningModel=Mistral-7B-v0.32026.0448.984.7
CoTModel=Qwen3-4B2026.0445.924.01
CoTModel=Llama-3.1-8B2026.0440.823.99
CoTModel=Mistral-7B-v0.32026.0436.734.09
Direct PromptingModel=Qwen3-4B2026.0432.654.22
Direct PromptingModel=Llama-3.1-8B2026.0428.574.16
Direct PromptingModel=Mistral-7B-v0.32026.0427.554.23
SBERT2026.0411.224.85