claim

active

claim:shot-midpoint-ordering-k50-b10-k50-b8-k50-b9-tracks-pretraining-exposure-density

Shot midpoint ordering k50(B10) < k50(B8) ≈ k50(B9) tracks pretraining exposure density

Interpretation that pattern density from pretraining determines few-shot requirements

Source paper

extracted_from

(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang

finding

B10 shot midpoint k50 = 0.28 ± 0.05 shots with accuracy 94.8 ± 1.2%
supports
Lowest threshold condition in E2; near-zero/one-shot threshold consistent with high pretraining density

community

Few-shot anchoring & latent structure
members_of
How minimal examples disambiguate and recruit latent arithmetic/reasoning interpretations in LLMs
Few-shot learning phase transitions in neural networks
members_of
Empirical characterization of k50 midpoints and transition widths across transformer models, tracking how pretraining density ρd/dr predicts in-context learning thresholds.
Few-shot arithmetic learning thresholds
members_of
k50 and phase width metrics for multi-base addition across bases 8, 9, and 10

cosine ≥ 0.65 · no typed edge

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

Hypothesis: Shot midpoint ordering k50(B10) < k50(B8) ≈ k50(B9) follows pretraining exposure densityhypothesis0.937
E2 prediction that bases with higher pretraining exposure require fewer shots to cross threshold
Shot midpoint ordering k50(B10) < k50(B8) ≈ k50(B9) and transition widths correlate with mismatch D(P0∥PT)hypothesis0.896
Testable prediction for Experiment 2
k50 ordering: B10 (0.28) < B8 (1.83) < B9 (2.91) follows pretraining densityfinding0.870
Monotone ordering consistent with k50 ∝ dr/ρd.
Shot midpoints follow k50 ∝ dr/ρd; higher cohesion and lower mismatch yield fewer required examplesclaim0.842
Core quantitative prediction of UCCT validated by E2 threshold ordering
The ordering of few-shot thresholds k50 and transition widths aligns with k50 ∝ dr/ρd.claim0.818
Interpretation of E2 results.
shot midpoint k50concept0.804
Number of in-context exemplars to reach 50% accuracy in E2.
Hypothesis: Peak alignment location S_max and normalized trajectory area AUS_N predict shot midpoints θ50hypothesis0.783
E3 prediction that internal geometry provides a bridge to behavioral thresholds
few-shot threshold (k50/θ50)concept0.760
Shot count needed to reach 50% accuracy; reflects when anchoring strength crosses critical value.