| paper | d4-journals-jet-2009-0098 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | yes |
Theorem 1 and the numbered lemmas give an axiomatic characterization, not a computational result. Winner determination over approval types is a well-defined, readily computed follow-on, but it does not supply the missing computational anchor from this paper.
fails bit a — no named computational result to mirror
The proposed winner query applies the characterized Size Approval Voting rule to a population distribution; it does not mirror a computational result. The paper's characterization and its proof lemmas remain axiomatic.
The strongest positive case is a natural population model for Size Approval Voting, but this paper supplies no named computational result to anchor it to. Its Theorem 1, proved in the paper, says that a family of voting rules satisfies Consistency in Voters, Anonymity, Neutrality, Contraction, and No-Veto if and only if there is a nonnegative, weakly decreasing weight vector \(w\) such that an alternative wins exactly when its weighted approval total is maximal. That is an axiomatic characterization. The theorem and the numbered lemmas do not establish a complexity result or an algorithm. So there is no qualifying anchor, and no lead among valid anchors.
A coherent follow-on question is Continuous Size-Approval Winner Determination. An instance has a candidate set \(X\), a list of \(\tau\) approval types \(S\subseteq X\) with positive rational masses \(\mu_S\) summing to one, and a rational weight vector \(w_1\ge\cdots\ge w_{|X|}\ge0\). For each \(x\in X\), define
\[
\mathrm{sc}_{\mu}(x)=\sum_{S:\,x\in S}\mu_S w_{|S|}.
\]
The task is to return the full winning set \(\arg\max_{x\in X}\mathrm{sc}_{\mu}(x)\). There is no strategic action to choose: the output is the set maximizing this score. A large electorate using a small menu of repeated approval ballots gives a plausible regime, with \(n\gg\tau\); each type is a ballot \(S\), and its mass is the fraction of the electorate casting it. An ordinary finite profile maps exactly to this representation by setting \(\mu_S=n_S/n\).
This is recognizable as the paper’s rule applied to a continuous population, and the winner query is directly solvable by summing the \(\tau\) contributions for each candidate. But that easy computation is a follow-on suggested by the rule, not a result the paper proves. The weak point is decisive under the requested standard: the paper’s contribution is its finite-electorate axiomatic characterization. Turning that into a continuous axiomatic question would fall outside the programme’s scope, while the natural computational query has no substantive complexity result here to mirror.
The strongest objection is that the paper has no computational result to mirror. Theorem 1 characterizes Size Approval Voting by axioms; Lemma 1 and the other numbered results are parts of that axiomatic analysis. None states a complexity bound, hardness result, or algorithm.
The proposed continuous winner-determination question is well-defined, and repeated approval ballots give it a sensible high-multiplicity interpretation. But with rational weights and masses, it amounts to computing
\[
\mathrm{sc}_{\mu}(x)=\sum_{S:\,x\in S}\mu_S w_{|S|}
\]
for each candidate and returning the maximizers. This takes \(O(m\tau)\) arithmetic operations. For a finite profile with \(n_S\) ballots of type \(S\), setting \(\mu_S=n_S/n\) divides every candidate’s score by the same \(n\), so the winning set is unchanged. That calculation adds no algorithmic question beyond evaluating the rule as defined. The paper does not present winner determination as a research problem; its result concerns which rules satisfy the axioms.
A continuous restatement of the characterization would still be axiomatic, outside ChoCo’s computational scope. A more ambitious question about mass bribery, manipulation, or winner robustness could be worthwhile, but it would need an intervention model—allowed ballot changes, costs, or objectives—that this paper does not supply. It would be a new problem suggested by the rule, rather than a continuous mirror of a result here.
I would not argue that approval ballots lack meaningful multiplicity: populations with many repeated ballot types are plausible, and the scoring formula carries over cleanly. The negative case is strong for this paper as a ChoCo anchor. It is weaker as a claim that no worthwhile continuous problem could ever be built from the paper’s subject matter.
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.