| paper | d4-journals-mss-2017-0005 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | yes |
Theorems 2–5 establish incentive properties, not computational results under the stated criterion. Computing the egalitarian allocation on a type-level flow model is a plausible follow-up, but it does not provide a continuous analogue of any anchored theorem.
fails bit a — no named computational result to mirror
The proposed anchors, Theorems 2–5, are all incentive results; the paper's max-flow-based allocation procedure is not stated as a numbered computational result. The proposed computational follow-up therefore covers no anchored theorem.
There is a plausible population mirror here, but the paper has no named computational result to anchor it. Under your criteria, that means I cannot make a qualifying positive case from this paper.
The sensible regime would be a large clearinghouse in which many suppliers and demanders share the same relevant attributes. A type would specify an agent’s side, per-person peak, bounds where applicable, and compatible types on the other side. Its mass \(\mu_t\) would be the fraction of the population with that complete description. The decision would be a flow \(z_{tu}\) between supplier and demander types; dividing by type mass gives each type’s per-person transfer. This makes population mass the continuous object, rather than relying on the paper’s already-fractional commodity flows. The scenario could be bulk allocation of aid or placements among recurring classes of participants.
The nearest named anchors are Theorem 2, proved here, which establishes peak group strategyproofness in the one-sided model: no coalition of suppliers can misreport peaks so that all its members weakly benefit and at least one strictly benefits. Theorem 3, also proved here, establishes the corresponding peak group strategyproofness in the two-sided model, where suppliers and demanders both report peaks. Theorem 4 establishes link group strategyproofness when a coalition consists only of suppliers or only of demanders; Theorem 5 combines link and peak reports under that same coalition restriction. These are strategic or axiomatic results. None asserts that an underlying computational problem is in P, NP-hard, FPT, or another complexity class. The paper describes a max-flow-based egalitarian algorithm, but does not give a numbered complexity result for computing it.
So there is no valid “one continuous problem per anchor” to hand to a prover under the requested rules. A natural *unanchored follow-up* would be: given rational masses of supplier and demander types, per-person peaks and bounds, and a compatibility relation between types, compute the type-level allocation selected by the egalitarian mechanism. It would likely be tractable through type-compressed lexicographic flow methods. Its weakness is decisive for this assignment: that computational question is not what Theorems 2–5 establish. Their main contribution is about incentives, and continuizing that contribution would continuize an axiomatic question, which is outside ChoCo’s stated scope.
The strongest case against a mirror is that none of the proposed anchors is a computational result. Theorem 2 proves peak group strategyproofness for suppliers in the one-sided model; Theorem 3 proves it for peaks on both sides; Theorem 4 handles link reports by single-side coalitions; and Theorem 5 combines reports under that same coalition restriction. Each states an incentive property. None classifies the complexity of a problem or gives an algorithmic bound. The paper’s main contribution is the group-strategyproofness proof, so a continuous version of that contribution would continuize an axiom—the programme explicitly excludes that.
There is a plausible high-multiplicity setting: a bulk rationing clearinghouse with recurring supplier and demander types, specified by side, per-person peak, bounds, and compatible types. A type-level flow gives a natural mass model. The already fractional commodity and the network’s compatibility constraints do not undermine that case. They make the allocation model natural; they do not turn the paper’s incentive theorems into computational results.
The best direct mirror of Theorems 2–5 would ask whether mass coalitions can profitably misreport peaks or links. That remains a question about group strategyproofness, even when coalitions and reports are represented by masses. Asking an algorithm to find a violating coalition changes the formulation into a search problem, but its witnesses are exactly violations of the same axiom. Computing the egalitarian allocation on a type-level network is a separate computational question; the paper describes a max-flow-based algorithm, but none of the anchored theorems establishes its complexity.
A stronger computational follow-up might measure the minimum mass of a *mixed-side* coalition that can manipulate links. That is plausible in a population model, but it moves beyond Theorem 4’s same-side restriction and adds an optimization objective the theorem does not study. It may be worth investigating on its own; it does not rescue any of the proposed anchors as continuous mirrors of the paper’s results.
So the negative case is strong for this assignment: the proposed anchors do not meet ChoCo’s computational scope. It is weaker as the universal claim that no worthwhile continuous project could arise from this model. The paper supports a sensible population regime, and a newly formulated algorithmic manipulation problem might be valuable; the paper’s stated results simply do not supply that case.
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.