ResearchBenchmarks20 Questions on 20Q S128Follow10.8Worst Case Interaction LengthGoT10.46412.7321517.268Feb 2, 2026Evaluation ResultsMethodMethodLinksWorst Case Interaction LengthGoTBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0210.8GoTBackbone=GPT 4.1Backbone=GPT 4.12026.0211.8UoTBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0212UoTBackbone=GPT 4.1Backbone=GPT 4.12026.0213DCBackbone=GPT 4.1Backbone=GPT 4.12026.0214.6DPBackbone=GPT 4.1Backbone=GPT 4.12026.0216.2DCBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0217.9DPBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0219.2