Pith. sign in

REVIEW 3 major objections 1 minor

In ZFC alone, b is strictly below cof(UN) and non(N) equals non(UN) but is strictly below cof(UN); under add(N)=c one gets cof(UN)=d_c, and a full four-way separation of the UN invariants is consistent.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

T0 review · grok-4.5

2026-07-15 02:11 UTC pith:IKECDYCZ

load-bearing objection Abstract-only ZFC inequalities and a four-way UN separation look like solid specialist progress, but nothing is checkable without the body. the 3 major comments →

arxiv 2607.12936 v1 pith:IKECDYCZ submitted 2026-07-14 math.LO

Cardinal invariants on universally null sets

classification math.LO MSC 03E1703E3503E15
keywords universally null setscardinal invariantscofinality of idealsnull idealbounding numberdominating numberCichoń diagramforcing
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper studies the four classical cardinal invariants (add, cov, non, cof) of the ideal of universally null sets of reals. Universally null sets are those that receive measure zero under every continuous Borel probability measure; they form a sigma-ideal UN properly containing the null ideal N. The author proves two ZFC inequalities that pin down how large the cofinality of UN must be: the bounding number b is always strictly smaller than cof(UN), and the uniformity of the null ideal equals the uniformity of UN yet is still strictly smaller than cof(UN). Under the extra assumption that the additivity of N equals the continuum, the cofinality of UN is identified exactly with the dominating number of the poset of functions from continuum to continuum. Finally, a forcing construction is given that separates all four UN-invariants at once. A sympathetic reader cares because these results place UN firmly inside the Cichoń diagram and show that the extra room between N and UN can be used to realize every inequality that the combinatorial definitions allow.

Core claim

In ZFC one has b < cof(UN) and non(N) = non(UN) < cof(UN). Assuming add(N)=c one has cof(UN)=d_c. The constellation add(UN)<cov(UN)<non(UN)<cof(UN) is consistent.

What carries the argument

The cofinality of the ideal UN of universally null sets, together with an adaptation of Yorioka's technique that, under the hypothesis add(N)=c, identifies cof(UN) with the dominating number d_c of ^c c.

Load-bearing premise

The exact equality cof(UN)=d_c requires the extra set-theoretic hypothesis that the additivity of the null ideal equals the continuum, and that Yorioka's combinatorial technique transfers to universally null sets without new obstacles.

What would settle it

A model of ZFC in which either b equals cof(UN), or non(N) is strictly smaller than non(UN), or (under add(N)=c) cof(UN) differs from d_c, or a proof that the four-way separation of the UN invariants is impossible.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 1 minor

Summary. The manuscript studies cardinal invariants of the ideal UN of universally null sets. It claims three principal results: the ZFC inequalities b < cof(UN) and non(N) = non(UN) < cof(UN); the conditional equality cof(UN) = d_c under the hypothesis add(N) = c, obtained by adapting Yorioka’s technique; and the consistency of the strict chain add(UN) < cov(UN) < non(UN) < cof(UN).

Significance. If the claimed theorems hold, the paper would substantially clarify the place of the universally null ideal among the classical cardinal characteristics of the continuum. The ZFC separations of cof(UN) from b and from non(N), the conditional identification with d_c, and a consistent four-way separation would put UN on a footing comparable to the null and meager ideals, and would be of clear interest to researchers in set theory of the reals.

major comments (3)
  1. [Abstract] The ZFC claims b < cof(UN) and non(N) = non(UN) < cof(UN) are announced as theorems, yet no proofs, lemmas, or intermediate constructions appear in the available text. These inequalities are load-bearing for the paper’s central contribution; without the body they cannot be checked for gaps or hidden selection principles.
  2. [Abstract] The equality cof(UN) = d_c is asserted under add(N) = c by an adaptation of Yorioka’s technique. The abstract gives no indication of which combinatorial obstacles arise in the adaptation or how they are resolved; that step is essential to the conditional identification and requires detailed verification.
  3. [Abstract] The consistency of add(UN) < cov(UN) < non(UN) < cof(UN) rests on an unspecified forcing construction. Preservation properties for the ideal UN and the mechanism that separates the four invariants cannot be assessed from the abstract alone.
minor comments (1)
  1. [Abstract] The abstract is clear and well-structured, but the notation d_c is not expanded; a brief parenthetical definition would aid readers outside the immediate subfield.

Circularity Check

0 steps flagged

No circularity detectable: abstract-only ZFC inequalities and consistency claim with no definitional reduction or fitted prediction visible.

full rationale

The available text is only the abstract. It asserts pure ZFC inequalities (b < cof(UN) and non(N)=non(UN)<cof(UN)), a conditional equality cof(UN)=d_c under the extra hypothesis add(N)=c obtained by adapting Yorioka's technique, and a consistency result for the four-way separation of the UN invariants. None of these statements exhibits a self-definitional loop, a fitted parameter renamed as a prediction, a load-bearing uniqueness theorem imported solely from the same authors, an ansatz smuggled via self-citation, or a renaming of a known empirical pattern. There are no equations, no parameter fits, and no internal derivation chain that can be reduced to its own inputs. Self-citation risk cannot be assessed without the body; under the hard rules an abstract-only pure-mathematics claim of this form receives score 0 with empty steps. The reader's score of 2 and the skeptic's note that the body is unavailable are consistent with this non-finding: absence of the body is a verification barrier, not circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

Abstract-only review. The work sits in ZFC set theory of the reals. Free parameters are not expected for pure inequalities; the conditional theorem uses the hypothesis add(N)=c. No new entities (particles, forces, etc.) appear. Axioms are the usual background of cardinal invariants and the definition of universally null sets.

axioms (3)
  • standard math ZFC set theory (including choice)
    All stated theorems and the consistency result are framed inside ZFC.
  • domain assumption Standard definitions of the ideals N (null) and UN (universally null) and of the cardinal invariants add, cov, non, cof, b, d, d_c
    The paper works with the classical Cichon-style characteristics of these ideals; the abstract assumes the reader knows them.
  • domain assumption add(N) = c (used for cof(UN) = d_c)
    Explicit extra hypothesis for the second main theorem; not proved in the paper.

pith-pipeline@v1.1.0-grok45 · 6025 in / 2144 out tokens · 24309 ms · 2026-07-15T02:11:05.830129+00:00 · methodology

0 comments
read the original abstract

We investigate the cardinal invariants on universally null sets. In particular, we prove $\mathfrak{b} < \operatorname{cof}(\mathcal{UN})$ and $\operatorname{non}(\mathcal{N}) = \operatorname{non}(\mathcal{UN}) < \operatorname{cof}(\mathcal{UN})$ in $\mathsf{ZFC}$. Also, assuming $\operatorname{add}(\mathcal{N}) = \mathfrak{c}$, we prove $\operatorname{cof}(\mathcal{UN}) = \mathfrak{d}_\mathfrak{c}$ by adapting Yorioka's technique. Moreover, we prove the consistency of $\operatorname{add}(\mathcal{UN}) < \operatorname{cov}(\mathcal{UN}) < \operatorname{non}(\mathcal{UN}) < \operatorname{cof}(\mathcal{UN})$.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.