| paper | d4-journals-ai-2017-0053 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | unclear |
Theorem 6 and Theorem 4 constrain which axioms fusion operators can satisfy, while the representation theorems characterize operators. The paper gives no algorithm, complexity claim, or computational formulation used to compute, so it fails bit (a). A continuous axiomatic analogue may be a coherent extension, but it cannot supply the required computational mirror.
fails bit a — no named computational result to mirror
No continuous mirror of a named computational result survives; the proposed population-level analogue of the axioms does not add a computational question.
I can’t give a qualifying positive case: the paper contains no named computational result of the kind the brief requires. Its results are about rational properties and representations of belief-fusion operators; it gives no complexity classification, algorithmic result, or running-time analysis.
The closest anchor is Theorem 6, proved in this paper: every ES basic fusion operator satisfying Standard Domain, Pareto, and Independence also satisfies the Dictator postulate. That is a substantive Arrow-style impossibility theorem, but it is not a result about the complexity of a computational problem. Theorem 4 is likewise an impossibility about representing epistemic states with propositional formulas while retaining specified properties. Neither can honestly anchor a tractability, hardness-transfer, or continuum-specific-hardness question.
The paper’s profiles—functions from finite agents to epistemic states—could motivate a continuous population model, with epistemic states as types and mass as population share. But asking whether a fusion operator on that population must be dictatorial would extend the paper’s axiomatic question, not mirror a named computational result. Under the requested anchor rule and ChoCo’s computational scope, there is no continuous problem here I can present as the required mirror.
There is no qualifying anchor to defeat: the proponent names none. The closest candidate, Theorem 6, is a theorem about axioms of fusion operators: Standard Domain, Pareto, and Independence imply Dictatorship. Theorem 4 is also axiomatic, and the representation theorems characterize operators rather than their computational complexity. The paper states no algorithmic problem, complexity classification, or running-time result. Under ChoCo’s stated scope, that is decisive.
The strongest proposed mirror would replace each finite profile with a distribution over epistemic-state types and ask whether continuous analogues of the axioms still force dictatorship. That is a coherent research question, but its result would still be axiomatic. The continuum changes the society’s representation; it does not turn the paper’s central question into a computational one. Defining a new task—such as deciding whether a given encoded operator satisfies those axioms—would require an encoding and computational problem absent from the paper. That would be a new project, not a mirror of a named result.
There is also a modeling obstacle to that rescue. The paper allows operators to depend on agent identity and order—for example, its projective operators select the last agent in a finite society. A distribution over epistemic states forgets that identity. One could put identity into types, but then the relevant agents need not have repeated types; or define a positive-mass decisive bloc, but that changes the dictatorship question. These are workable choices for a new axiomatic study, not grounds for a computational mirror.
So the negative case is strong for the programme’s decision: this paper offers no paper-anchored computational mirror. It cannot establish that no worthwhile project could ever be built from these ideas; it does establish that such a project would need a new computational question rather than continuizing one of this paper’s 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.