List your company

Information Seeking on 20Q Breeds weighted (test)

47.8Worst-case Weighted PayoffDP
22.21628.85835.542.142Feb 2, 2026

Evaluation results

MethodLinks
DPBackbone LLM=Qwen 2.5 72B Instruct2026.0247.8
UoTBackbone LLM=Qwen 2.5 72B Instruct2026.0247
DCBackbone LLM=Qwen 2.5 72B Instruct2026.0245.7
DCBackbone LLM=GPT 4.12026.0241.1
DPBackbone LLM=GPT 4.12026.0236.9
GoTBackbone LLM=Qwen 2.5 72B Instruct2026.0232.3
UoTBackbone LLM=GPT 4.12026.0232.1
GoTBackbone LLM=GPT 4.12026.0223.2