Effective Frontiers: A Unification of Neural Scaling Laws

Jiaxuan Zou, Zixuan Gong, Ye Su, Huayi Tang, Yong Liu

arXivarxiv/2602.02593Feb 1, 2026cs.LGscore 0

Neural scaling laws govern the prediction power-law improvement of test loss with respect to model capacity ($N$), datasize ($D$), and compute ($C$). However, existing theoretical explanations often rely on specific architectures or complex kernel methods, lacking intuitive universality. In this paper, we propose a unified framework that abstracts general learning tasks as the progressive coverage of patterns from a long-tail (Zipfian) distribution. We introduce the Effective Frontier ($k_\star$), a threshold in the pattern rank space that separates learned knowledge from the unlearned tail.…

SourceJSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2602.02593
curl -X POST /api/v1/records/arxiv/2602.02593/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2602.02593/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'