finding
active
finding:with-raw-data-wider-token-strides-give-more-accurate-approximations-at-same-token-budgetWith raw data, wider token strides give more accurate approximations at same token budget.
Cost tradeoff.
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
- Wider stride improves raw accuracy at fixed budget because averaging over a window reduces variance.supportsExplanation of raw behavior.
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.
- With smoothed data, spacing and S trace the same Pareto frontier of accuracy vs token cost.finding0.818Cost tradeoff.
- With raw data, wider spacing has little effect on accuracy except at high S and near forking points.finding0.813Raw data behavior.
- §5.5.
- Design choice validated empirically.
- Key result demonstrating advantage of stepwise over all-token steering strategy
- Fig. 7.
- Justifies the choice of response tokens for persona vector extraction in the pipeline
- Smoothed behavior.