VS-Standard achieves KL divergence of 0.54 from pretraining distribution on Open-Ended QA, vs. 3.14 for Direct and 0.58 for Sequence

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• VS-Standard achieves KL divergence of 0.12-0.13 from pretraining distribution on US state naming task, vs. 14.89-16.16 for Direct promptingfinding0.933
  Strong empirical evidence that VS recovers pretraining distribution while direct prompting collapses
• VS-Standard achieves KL divergence of 0.027 from uniform distribution on dice roll simulation, vs. 0.926 for Direct promptingfinding0.874
  Shows VS enables LLMs to better approximate random behavior compared to direct prompting
• VS-Multi achieves Coverage-N of 0.71 on Open-Ended QA vs. 0.10 for Direct, while maintaining precision of 0.96finding0.803
  Demonstrates VS generates a broader range of valid answers without sacrificing accuracy
• VS-Standard achieves human-rated diversity scores of 2.39/3.06/3.01 for poem/story/joke vs. 1.90/2.74/1.83 for Direct prompting on 4-point scalefinding0.789
  Human study confirming automatic diversity metrics align with human perceptions
• VS-Multi achieves average accuracy of 37.5% on math benchmarks (avg of MATH500, OlympiadBench, Minerva) with Gemini-2.5-Flash as generator and Qwen3-4B as SFT model, vs. 40.7% baselinefinding0.781
  Best VS result in synthetic data generation for math, demonstrating downstream improvement through diversity
• VS recovers 66.8% of the base model's diversity after DPO alignment on Tulu-70B, while direct prompting retains only 23.8%finding0.780
  Quantifies how much of the base model's diversity VS can recover compared to baseline prompting
• VS-Standard achieves 1.86x diversity gain at only 1.12x cost increase over Direct prompting on poem generationfinding0.773
  Cost-diversity trade-off analysis showing VS's practical efficiency
• All VS variants maintain refusal rates above 97% on StrongReject benchmark, within 0.3-0.8 percentage points of the Direct baselinefinding0.771
  Confirms VS does not compromise safety alignment while improving diversity