| paper | d4-journals-mss-2022-0002 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | yes |
Neither named theorem asserts an algorithm, complexity bound, or hardness result; both classify normative properties of allocation rules. The type-mass Compromise Convex allocation is a plausible, precise computational question, but it is introduced by the analyst rather than anchored in a computational result of the paper.
fails bit a — no named computational result to mirror
The proposed extension carries the Compromise Convex allocation rule to claimant types with masses; it does not turn Theorem 1 or Theorem 2’s normative property classifications into computational results.
The strongest honest case is that a continuous-population version of the paper’s claims problem is natural and precise. But the paper has no eligible computational anchor: none of its numbered results says that a problem is in P, NP-hard, FPT, or otherwise computationally classified.
The closest results are Theorem 1 and Theorem 2, both proved in this paper, but both are about normative properties. Theorem 1 says the fixed-parameter piece-wise mixture preserves equal treatment of equals, order preservation, anonymity, continuity, resource monotonicity, claim monotonicity and consistency; with resource-monotonic component rules, it also preserves population monotonicity. It says the mixture need not preserve minimal rights first, self-duality, or composition up or down. Theorem 2 says the Compromise Convex rule satisfies equal treatment of equals, order preservation, anonymity, continuity, resource monotonicity, self-duality and population monotonicity; it satisfies claim monotonicity when the compromising degree exceeds \(0.5\), and fails it below \(0.5\). It also fails minimal rights first, composition up and down, and consistency. These are named results worth carrying into a population model, but they do not meet the requested computational-anchor standard.
A plausible mirror would be continuous Compromise Convex allocation. Take \(\tau\) claimant types, where type \(t\) has claim \(c_t\), and a rational mass fraction \(\mu_t>0\), with \(\sum_t\mu_t=1\). Let \(e\) be the estate per claimant, with \(0<e<\sum_t\mu_t c_t\). The decision is an award \(a_t\) per unit of type-\(t\) mass; awards must satisfy \(0\le a_t\le c_t\) and \(\sum_t\mu_ta_t=e\). The rule prescribes the unique allocation, so there is no separate optimization objective. Specifically, set \(\theta=e/\sum_t\mu_tc_t\); find \(\lambda,\eta\) satisfying
\[
\sum_t\mu_t\min\{\lambda,c_t\}=e,
\qquad
\sum_t\mu_t\max\{0,c_t-\eta\}=e;
\]
then output
\[
a_t=\theta\min\{\lambda,c_t\}+(1-\theta)\max\{0,c_t-\eta\}.
\]
This is the paper’s CC rule extended to masses: its CEA and CEL components are computed with type masses in the aggregate-award equations.
The regime is a large pool of claimants governed by a small number of standardized claim amounts—for example, a mass-market insolvency with claims set by a few contract tiers. Then \(N\), the number of claimants, can be large while \(\tau\), the number of distinct claim types, stays small. Claim amount is a complete type for the distinctions this rule uses. An exact computation question for this mass allocation should be tractable in \(\tau\) and the rational input length, since its thresholds can be found by sorting the claim amounts. That is a reasonable new computational question based on the paper’s rule, but it is not a result established by the paper.
The connection to the authors’ work is strongest at the model level: claims are already divisible, and replacing a finite list of claimants with masses of identical claimants preserves the claims problem and its CC allocation rule. Theorem 2’s population-monotonicity result also makes the extension conceptually apt, although proving a mass analogue of that axiom would be an axiomatic question rather than the computational contribution sought here.
The weak point is decisive for this task: the paper studies solution properties, not the complexity of computing allocations. The proposed mass version is a credible high-multiplicity extension, but it does not turn any of the paper’s named results into a hardness-or-tractability anchor. There is therefore a plausible mirror to explore, but no qualifying anchor on which to build the requested continuous complexity case.
The strongest objection is that this paper has no computational result to mirror. Theorem 1 and Theorem 2 classify normative properties of allocation rules. Neither studies a decision or optimization problem, nor asserts a complexity bound. They are the only anchors the proponent identifies, so neither survives the programme’s computational gate.
The mass version of Compromise Convex allocation is nevertheless a sensible model. For a large pool of claimants with a small number of claim amounts, type masses preserve the inputs the rule uses. The claims are already divisible, but that is continuity of the *outcome*, not of the population; it neither makes this prior continuous-population work nor rules out a mass model. Nor does claimant identity supply a convincing objection: for these rules, people with the same claim are interchangeable.
That leaves the proponent’s proposed task: compute the allocation from type masses. It is a legitimate task one could formulate, but it is not a result in the paper. The paper defines the allocation rules through CEA and CEL thresholds and studies their properties. Turning those definitions into a complexity problem about weighted inputs would add a computational agenda the authors did not pursue; it would not mirror a computational claim they made. Theorem 1’s preservation results could likewise be restated for mass populations, and Theorem 2’s population-monotonicity property is the most natural candidate: one could compare outcomes before and after adding positive mass of a claimant type. But those are still axiomatic questions, not complexity results.
A richer computational problem—such as choosing claims or mixture parameters to meet a policy target—would require introducing a new objective or intervention. Nothing in the paper selects one, so it would be a new research direction rather than a continuous mirror of its results. The negative case is therefore strong under ChoCo’s stated scope, though not as a universal claim about modelling: the mass allocation itself is plausible, and its axiomatic study could be worthwhile outside that scope.
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.