finding
active
finding:at-s-200-variation-across-runs-tvd-0-0451-nearly-equals-sampling-variation-within-a-run-tvd-0-0442At S=200 variation across runs (TVD 0.0451) nearly equals sampling variation within a run (TVD 0.0442).
Supports multinomial model.
Source paper
extracted_from(2026) · Eric Bigelow · Amir Zur · Satchel Grant · Tal Haklay +6
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Abstract claim.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Accuracy metric versus S=200 reference.
- Sampling noise decreases in proportion to 1/√S, giving a linear relation between log TVD and log S.hypothesis0.815Derived from multinomial sampling.
- Log-log slope of −0.4903 of pooled replicate TVD vs S through S=200 on two development questions, near −1/2.finding0.791Noise scaling.
- Forks missed with coarse spacing.
- Shows VS substantially better approximates the pretraining distribution than baseline methods
- Shows VS enables LLMs to better approximate random behavior compared to direct prompting
- Noise matches null.
- Strong empirical evidence that VS recovers pretraining distribution while direct prompting collapses