AWhat changed
Model-training tradeoffs could be summarized and forecast approximately within measured regimes.
Scientific scope
2020 ten-author paper, cross-entropy loss and tested model/data ranges; not universal laws of intelligence or guaranteed compute-optimality elsewhere.
Evidence
The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. These are empirical relationships for the architectures, data and loss regimes tested in the 2020 paper, not universal laws of intelligence or guarantees outside the measured ranges.
Contributors
- The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; this collective display does not merge person identities or measure labor shares.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction