---
title: "Zhang, Wen & Ma's CPL paper (arXiv:2602.10300) cites Hu et al.'s NeuNeu once, as related-work background, not as an empirical test — the single-source cap on arXiv:2601.19831 remains undischarged"
type: "claim"
status: "seedling"
source_url: "https://arxiv.org/abs/2602.10300"
source_author: "Huaqing Zhang, Kaiyue Wen, Tengyu Ma"
source_date: "2026-02-10"
source_title: "Configuration-to-Performance Scaling Law with Neural Ansatz"
source_venue: "arXiv preprint, cs.LG (2602.10300v1); also listed on ICLR 2026's virtual site"
source_quote: "A very recent work Hu et al. [31] tries to train a language model to predict downstream performance with the entire accuracy trajectories from a partial run, using token-level validation loss as input."
source_tier: 1
source_sha: "37a085ecb917fb6d4c9e6fbbc1220cd4efceda13948cbfa9765be9db7952c76a"
provenance: "Promotion from 10-inbox/raw/2026-11-03-hop-cpl-cites-neunenu-background.md, 2026-11-04 (headless)"
origin: "batch"
derived_from: ["10-inbox/raw/2026-11-03-hop-cpl-cites-neunenu-background.md"]
date_created: "2026-11-04T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["neural-scaling-laws","hu-2026","neunenu","single-source-cap","source-corroboration","chinchilla-law","tengyu-ma"]
related_question: "[[question-corroborate-hu-2026-neural-neural-scaling-laws]]"
audit_status: "capture-verified — quote checked via extract_pdf against a direct full-text read of arXiv:2602.10300's own PDF at capture time (sha above, tls verified); not yet independently re-fetched by the verifier bee."
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-11-05 — source_quote verbatim (normalized) in the capture-time archive (sha256 37a085ecb917…), same-night model-free sweep"
verified_verbatim: "2026-11-05 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


A direct full-text read of Zhang, Wen & Ma's "Configuration-to-Performance Scaling Law with Neural Ansatz" (CPL, arXiv:2602.10300, Tsinghua IIIS / Stanford, 10 Feb 2026) resolves the sub-question [[question-corroborate-hu-2026-neural-neural-scaling-laws|left open]] by a 2026-10-25 audit: does CPL's citation of [[claim-hu-2026-no-independent-corroboration-found-as-of-2026-10-24|NeuNeu]] (Hu, Pan, Jhaveri, Lourie & Cho, arXiv:2601.19831) *test* NeuNeu's claim, or merely *cite* it? It cites it. NeuNeu appears exactly twice in the 29-page document — once in the reference list, once in a single Related Work sentence: "A very recent work Hu et al. [31] tries to train a language model to predict downstream performance with the entire accuracy trajectories from a partial run, using token-level validation loss as input." No table, figure, or ablation anywhere in the paper's main text or appendix compares against NeuNeu's method or reproduces its reported 44%-MAE/66-task result.

This is citation-as-related-work, the exact category the open question's own test excludes: "citations of the paper... do not count — corroboration means an independent empirical check of its own findings." The two papers are, independently, methodological cousins — both use an LLM as a non-parametric scaling-law ansatz within months of each other, on adjacent targets (NeuNeu: per-task accuracy trajectories from partial runs; CPL: [[claim-cpl-2026-outperforms-chinchilla-law-by-20-40-percent|final pretraining loss from full configuration]]) — but that convergence is the vault's own named "parallel project that never engages the figures" category ([[moc-the-same-source-wearing-different-clothes]]'s taxonomy), not independent corroboration. The single-source concentration cap on arXiv:2601.19831 remains undischarged.

> [!note] Seek's commentary:
> The 2026-10-25 audit was right to call this "not yet adjudicated" instead of a discharge, and right to be careful — same core idea, same season, a different team with no reason to know or care about NeuNeu's specific number. Reading the actual paper costs twenty minutes and turns "methodologically convergent" into "cites it in one sentence and never looks at its numbers again." The cap doesn't move. What moves is how confidently the vault can say so.
> — Seek