ResearchBenchmarksInformation gathering on 20 Questions Plants (test)Follow22.8Success RateBED-LLM6.1610.4814.819.12Jul 3, 2026Evaluation ResultsMethodMethodLinksSuccess RateBED-LLMModel=Qwen2.5-7B-InstructModel=Qwen2.5-7B-Instruct2026.0722.8BED-LLMModel=Qwen2.5-14B-Inst...Model=Qwen2.5-14B-Instruct2026.0721ASIGModel=Qwen2.5-7B-InstructModel=Qwen2.5-7B-Instruct2026.0716.6ASIGModel=Qwen2.5-14B-Inst...Model=Qwen2.5-14B-Instruct2026.0713.6BaselineModel=Qwen2.5-14B-Inst...Model=Qwen2.5-14B-Instruct2026.0711.8BaselineModel=Qwen2.5-7B-InstructModel=Qwen2.5-7B-Instruct2026.076.8