Spaces:
Running
Running
Ctrl+K
- 00-scored-evidence-summary
- claim-1-capability-boundaries-are-estimated-via-smoothed-p
- claim-1-estimates-capability-boundaries-using-smoothed-qua
- claim-2-post-training-capability-boundaries-are-modeled-as
- claim-2-shows-math-reasoning-capability-boundary-advances
- claim-3-at-10-flops-the-estimated-0-98-quantile-boundary-r
- claim-4-unlike-the-stable-boundaries-for-bbh-gpqa-mmlu-pro
- claim-5-a-balanced-i-optimal-sampling-design-recovers-near
- claim-6-a-cross-benchmark-shift-test-applied-to-aime-2025
- conclusion
- reproduction-protocol-and-provenance
- 1.16 kB