claim
active
claim:the-advantage-of-larger-models-is-not-merely-sample-efficiency-but-a-more-fundamental-limitation-of-smaller-models-under-infinite-dataThe advantage of larger models is not merely sample efficiency but a more fundamental limitation of smaller models under infinite data
Distinguishes from undertraining.
Source paper
extracted_from(2026) · Jing Huang · Daniel Wurgaft · Rachit Bansal · Laura Ruis +5
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Core phenomenological argument.
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.
- Core framing.
- Motivating question.
- Cost of removing superposition.
- Inverse scaling interpretation.
- Implication of PRH for AI fairness and bias
- Bigger models are more likely to converge to a shared representation than smaller modelshypothesis0.786Selective pressure toward convergence via model capacity
- Practical implication.
- Hypothesis box in §3.2.