ResearchBenchmarksPrompt Generation on DecTest prompt_gen 1000 samples no_hdsFollow0.932Spearman Rhoa_n-0.79336-0.345430.10250.55043Jun 1, 2026Evaluation ResultsMethodMethodLinksSpearman Rhoa_nmodel=Qwen2.5-3B, perm...model=Qwen2.5-3B, permutations=252026.060.932distinct-n2026.060.917BERTScore2026.060.878cos-sim2026.060.873C x a_nmodel=Qwen2.5-3B, perm...model=Qwen2.5-3B, permutations=252026.060.847SentBERT2026.060.747Cmodel=Qwen2.5-3B, perm...model=Qwen2.5-3B, permutations=252026.06-0.727