finding
active
finding:with-smoothed-data-spacing-and-s-trace-the-same-pareto-frontier-of-accuracy-vs-token-costWith smoothed data, spacing and S trace the same Pareto frontier of accuracy vs token cost.
Cost tradeoff.
Source paper
extracted_from(2026) · Eric Bigelow · Amir Zur · Satchel Grant · Tal Haklay +6
Neighborhood — ranked by edge-count
Papers (1)
paper
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 raw data, wider token strides give more accurate approximations at same token budget.finding0.818Cost tradeoff.
- Smoothed behavior.
- With raw data, wider spacing has little effect on accuracy except at high S and near forking points.finding0.802Raw data behavior.
- Choosing token indices to sample adaptively, as in optimal experiment design, could be more efficient than post-hoc smoothing.hypothesis0.775Future work.
- Fig. 7.
- Appendix B.
- Combining spaced resampling with smoothing cuts total budget by 1/8 with small error increase.finding0.762Main result 3.
- §5.5.