| paper | d4-journals-geb-2020-0090 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | medium |
| authors would recognise it | unclear |
The cases identify no named result that states a computational task: Theorem 1 gives a constructive choice-function characterization, but does not claim an algorithm or complexity result. A type-symmetric continuum version would need a new input model and computational question, and neither case specifies one as a mirror of a result in the paper.
fails bit a — no named computational result to mirror
Theorem 1's constructive characterization, Proposition 1's cited mechanism results, and the domain and embedding corollaries are not anchored to a computational result in this paper.
This paper contains no named computational result of the kind required for an anchor. Proposition 1, cited from earlier work, describes what deferred acceptance and cumulative offer achieve under substitutability and the law of aggregate demand. It does not establish a complexity classification. Theorem 1 constructs an equivalent choice function with stronger properties, but does not analyze the complexity of computing it. The other numbered results characterize mechanism properties, domains, and an embedding.
A continuous-population matching model could be proposed, but it would not mirror a named computational result of this paper. Under the stated rules, there is no qualifying anchor and therefore no continuous problem to specify per anchor.
The proponent raises no named result as an anchor, and I would keep that objection. The paper’s numbered results concern stability, strategy-proofness, equivalence of choice functions, and maximal domains. Theorem 1 is constructive, but it does not analyze the complexity of computing its construction. Proposition 1 reports earlier mechanism results. Deferred acceptance and cumulative offer are algorithms in the paper, but the paper gives no complexity classification for them.
The strongest alternative anchor is Theorem 1: ask whether a high-multiplicity algorithm can test the conditions or compute the equivalent virtual choice function. But the paper allows an arbitrary college choice map \(Ch_c:2^{X_c}\to2^{X_c}\), with contracts indexed by individual students. It specifies no succinct representation or oracle model for that map. A runtime bound in \(\tau\), the number of student types, therefore needs additional assumptions about how college choices treat those types. Those assumptions might yield a worthwhile computational problem, but they would define a new problem and input model; the theorem itself does not supply one to mirror.
The maximal-domain corollary and the stability and group-strategy-proofness results offer another route: characterize their analogues for a continuum of students. That could be interesting mechanism-design theory, but it would be an axiomatic question, outside ChoCo’s computational focus. A large cohort of students with shared preferences and treatment by colleges is a plausible high-multiplicity regime, so I would not argue that matching cannot be continuized. The difficulty is that the paper does not study computation in that regime. Even its strategy-proofness claims need care in an atomless model: a single student has zero mass, while requiring a positive-mass coalition to gain changes the deviation being studied.
So the case against is strong for excluding this paper as a ChoCo mirror: there is no named computational result to anchor, and the best rescue requires adding a computational question and its representation assumptions. It is weaker as the literal claim that no continuous matching project could be worthwhile. A type-symmetric, succinctly represented version of deferred acceptance might be worth studying, but it would be a new project prompted by the paper, not a continuous mirror of one of its results.
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.