hypothesis
active
hypothesis:choosing-token-indices-to-sample-adaptively-as-in-optimal-experiment-design-could-be-more-efficient-than-post-hoc-smoothingChoosing token indices to sample adaptively, as in optimal experiment design, could be more efficient than post-hoc smoothing.
Future work.
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.
- §5.5.
- Justifies the choice of response tokens for persona vector extraction in the pipeline
- With smoothed data, spacing and S trace the same Pareto frontier of accuracy vs token cost.finding0.775Cost tradeoff.
- Averaging acts as a low-pass filter that smooths away per-token detail and retains broad conceptsclaim0.773Intuition.
- Design choice validated empirically.
- Motivates nested SAE.
- Poisoning results.
- Honest limitation.