| paper | d4-journals-mss-2025-0056 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
The paper's numbered results are axiomatic impossibility claims about classification aggregation. None asserts an algorithm, a complexity result, or another computational claim, so the required computational anchor is absent.
fails bit a — no named computational result to mirror
Theorem 1 and Propositions 2–5 establish impossibility or existence claims about axioms; none supplies a computational result for the mirror to cover.
There is no named computational result in this paper to anchor a continuous mirror. Theorem 1, proved here, says that for \(m\ge p\ge2\), no classification aggregation function satisfies both Unanimity and Minimal Expertise. Propositions 2–5 give further impossibility results about axioms on finite-agent aggregation. None asserts a complexity classification or gives an algorithm for a computational problem.
The most plausible continuous setting would be a large classification panel represented by masses of agents with the same classification and expertise profile. But Minimal Expertise gives decisive power to particular individuals. A mass distribution either loses those identities or requires replacing individual decisiveness with a new, group-level condition; the paper’s theorems do not establish results about that replacement. And the resulting question would be axiomatic, rather than the computational question this programme asks for.
So the honest positive case is limited: large classification panels could make a continuous population model sensible, but this paper supplies no qualifying anchor, and I cannot state a continuous problem as the continuous form of one of its named computational results.
The decisive problem comes before the population limit: there is no computational result here to mirror. Theorem 1 and Propositions 2–5 are claims about whether an aggregator can satisfy combinations of axioms. They give no algorithmic problem, complexity bound, or optimization task. A distributional version of Theorem 1 could ask whether a continuous-population aggregator can be unanimous and expert, but that remains an axiomatic existence question—the kind of continuization ChoCo explicitly puts out of scope. Turning it into a task such as computing an outcome would require adding a rule-selection or optimization objective the paper does not supply.
The strongest proposed rescue is a large panel whose types encode complete classifications and expertise profiles. But Theorem 1’s expertise condition gives power to particular individuals. In an atomless population, one individual has zero mass and cannot determine an object’s classification; giving that power to a positive-mass expert cohort changes individual decisiveness into group decisiveness. Keeping named experts as atoms preserves their role, but leaves a hybrid finite-expert model rather than a continuous-population mirror of the theorem. Either way, the result one would study is still axiomatic. Large classification panels may be a sensible high-multiplicity setting, but they do not give this paper a worthwhile mirror for ChoCo’s computational programme.
The adversarial triple: the proponent anchors on up to three named results; the opponent sees that case and must defeat every anchor; the judge decides which case convinced it. These are the pipeline’s own outputs, generated by tools/triple_run.py — no human edited them. The paper’s own text is not reproduced here beyond the quoted statement above.