Spaces:
Running
Running
File size: 3,257 Bytes
6e3a665 9fd3976 6e3a665 9fd3976 6e3a665 9fd3976 6e3a665 9fd3976 6e3a665 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 | {
"schema_version": 1,
"title": "Repro - Prescriptive Scaling Reveals the Evolution of Language Model Capabilities",
"emoji": "\ud83d\udd2c",
"space_id": "bertfil/IkjsHRpuYY",
"paper": {
"arxiv_id": "2602.15327",
"openreview_id": "IkjsHRpuYY",
"title": "Prescriptive Scaling Reveals the Evolution of Language Model Capabilities",
"url": "https://arxiv.org/abs/2602.15327"
},
"tags": [
"icml2026-repro",
"paper-IkjsHRpuYY"
],
"updated_at": "2026-07-20T03:52:38+00:00",
"root": {
"slug": "index",
"title": "Repro - Prescriptive Scaling Reveals the Evolution of Language Model Capabilities",
"file": "pages/index.md",
"children": [
{
"slug": "00-scored-evidence-summary",
"title": "00 - Scored evidence summary",
"file": "pages/00-scored-evidence-summary/page.md",
"children": []
},
{
"slug": "claim-1-capability-boundaries-are-estimated-via-smoothed-p",
"title": "Claim 1 - Capability boundaries are estimated via smoothed pinball-los\u2026",
"file": "pages/claim-1-capability-boundaries-are-estimated-via-smoothed-p/page.md",
"children": []
},
{
"slug": "claim-2-post-training-capability-boundaries-are-modeled-as",
"title": "Claim 2 - Post-training capability boundaries are modeled as sigmoid f\u2026",
"file": "pages/claim-2-post-training-capability-boundaries-are-modeled-as/page.md",
"children": []
},
{
"slug": "claim-3-at-10-flops-the-estimated-0-98-quantile-boundary-r",
"title": "Claim 3 - At 10\u00b2\u2074 FLOPs, the estimated 0.98-quantile boundary reaches \u2026",
"file": "pages/claim-3-at-10-flops-the-estimated-0-98-quantile-boundary-r/page.md",
"children": []
},
{
"slug": "claim-4-unlike-the-stable-boundaries-for-bbh-gpqa-mmlu-pro",
"title": "Claim 4 - Unlike the stable boundaries for BBH, GPQA, MMLU-Pro, and MU\u2026",
"file": "pages/claim-4-unlike-the-stable-boundaries-for-bbh-gpqa-mmlu-pro/page.md",
"children": []
},
{
"slug": "claim-5-a-balanced-i-optimal-sampling-design-recovers-near",
"title": "Claim 5 - A balanced I-optimal sampling design recovers near-full-data\u2026",
"file": "pages/claim-5-a-balanced-i-optimal-sampling-design-recovers-near/page.md",
"children": []
},
{
"slug": "claim-6-a-cross-benchmark-shift-test-applied-to-aime-2025",
"title": "Claim 6 - A cross-benchmark shift test applied to AIME-2025 scores fin\u2026",
"file": "pages/claim-6-a-cross-benchmark-shift-test-applied-to-aime-2025/page.md",
"children": []
},
{
"slug": "reproduction-protocol-and-provenance",
"title": "Reproduction protocol and provenance",
"file": "pages/reproduction-protocol-and-provenance/page.md",
"children": []
},
{
"slug": "conclusion",
"title": "Conclusion",
"file": "pages/conclusion/page.md",
"children": []
}
]
},
"agent_view_tokens": 6563,
"revision": "1784519558058211500"
} |