| paper | d4-conf-atal-2015-0331 |
| 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 reports empirical patterns in voting behavior, but it has no numbered result asserting an algorithm, complexity bound, or other computational claim. Its convergence guarantee is attributed to prior work, and a continuous computational question would need an added update or optimization model. The plausible large-population poll setting does not supply the missing computational anchor.
fails bit a — no named computational result to mirror
The paper's empirical findings on poll voting, iterative voting, and convergence are not covered by a computational mirror.
I can’t make a valid anchor-based positive case from this paper. It contains no numbered theorem, lemma, corollary, or proposition asserting a computational complexity result. Its results are empirical findings about voting behavior. Table 3, for example, reports convergence rates in committees of 3, 5, and 7 voters; it does not establish an algorithmic result. The MBR convergence claim is attributed to prior work [19], so it is not an anchor proved by this paper.
The closest possible follow-on mirror would take the six rankings over three candidates as types, let \(\mu_t\) be each type’s population share, and track \(x_{t,c}\), the mass of type \(t\) currently voting for candidate \(c\). One could ask whether a specified mass-update version of myopic best response converges, or which equilibria it can reach. But the paper’s MBR behavior depends on individual voters being able to affect the outcome; an infinitesimal voter in a continuum has no such effect. Giving types the power to move a positive mass, or adding another mechanism for pivotality, would require a new model the paper does not study.
So that is a possible research direction, but not a continuous mirror anchored in a named result of this paper. There is no honest way to state the required problem per anchor here.
The strongest case against a mirror is that this paper has no computational result to mirror. Its findings are empirical: measured rates of compromise, herding, best-response moves, and convergence in experiments. Table 3 reports what happened in groups of 3, 5, and 7; it is not a complexity theorem. The convergence guarantee the paper mentions comes from prior work [19]. A new complexity result about iterative voting might be worthwhile, but it would not be a continuous mirror of a computational result in this paper.
The proposed MBR mirror has a deeper problem. In the committee experiment, a voter can change the winner by changing one ballot. In a nonatomic society, a single voter has zero mass and cannot change the outcome, so the paper’s notion of a pivotal MBR disappears. Letting an entire type’s mass switch together would restore consequential moves, but it changes the process from sequential individual updates to coordinated block updates. One could instead define a mean-field update rule, but that would be a newly specified dynamic whose convergence or equilibrium properties are not the paper’s empirical findings.
The better candidate is the poll experiment: online voters can form a large population, and the six preference rankings could be types with masses \(\mu_t\). That is a sensible high-multiplicity regime. But the paper studies how a *single person* responds to poll information, including herding and interpersonal differences. A continuum could record aggregate response rates by type, perhaps with behavioral disposition included in the type. That would support empirical estimation or prediction; it would not turn the paper’s findings into a computational question. To ask a complexity question, one would have to add a policy model, an optimization task, or a transition rule that the paper does not supply. Such a project may be valuable, but it would be a new computational agenda rather than this paper’s continuous mirror.
So I would not argue that no plausible continuous society exists: the poll setting has one. The stronger negative is narrower. Under ChoCo’s requirement that the mirror ask computational questions, this paper supplies no anchor that survives. Its natural continuizations either study aggregate human behavior empirically or replace the finite game with a newly defined mean-field dynamic. Neither mirrors a named computational result here.
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.