| paper | d4-conf-aaai-2020-0163 |
| 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 contains no numbered theorem, lemma, corollary, or proposition asserting an algorithmic or computational result, so bit (a) fails. Its abstract aggregation map and research agenda do not specify a rule or computational task; a continuous annotation problem would add a target rather than mirror a result of this paper.
fails bit a — no named computational result to mirror
There is no paper-authored computational result to mirror; the paper presents a general aggregation framework, surveys applications, and proposes research directions.
The paper has a plausible high-multiplicity setting: in crowdsourced annotation, a type could be a worker’s annotation, mass its share of the workforce, and the output a collective annotation. But the paper does not define a computational problem for that setting with a specified objective or decision criterion.
More decisively, it contains no numbered theorem, lemma, corollary, or proposition asserting a complexity or algorithmic result. The formal map \(F:L^n\to L\) and the discussion of complexity theory are part of a position paper’s framing, not a named computational result. Results mentioned from other authors are not anchors from this paper. So there is no honest anchor for a prover-ready continuous question or an expected tractability class; constructing one would require adding a problem the paper does not specify.
The strongest case against a mirror is that this paper supplies no computational result to mirror. It is a position paper: it defines aggregation abstractly, surveys application areas and contributing disciplines, and proposes a research agenda. It gives no numbered theorem, lemma, corollary, or proposition about the complexity or algorithms of a specified computational problem. Its discussion of complexity theory names a kind of analysis the field should use; it does not report such an analysis.
Crowdsourced annotation is a plausible high-multiplicity setting, and repeated annotation types could be a sensible model. But that repairs only the population side. The paper does not specify which aggregation rule to compute, what decision problem to ask about that rule, or what objective selects one output over another. The formal map \(F:L^n\to L\) leaves \(F\) arbitrary. Asking for its complexity has no determinate meaning without choosing and encoding a particular rule or a problem of designing one. A stronger annotation model—say, one with an explicit loss, worker-reliability assumptions, or constraints on the output—could yield a worthwhile continuous-population problem, but those choices would supply the computational question rather than mirror one in this paper.
The other tempting anchors fare no better: PageRank, peer grading, judgment aggregation, and the other named results are cited as work by other authors, not established here. They may motivate separate mirrors of those papers, but they cannot provide this paper with a result it does not contain.
So the negative case is strong for the claim that this paper has no paper-specific continuous mirror: changing the scenario or type space cannot create a missing computational target. It is weaker as a claim that no worthwhile ChoCo problem could be developed from the paper’s agenda. Crowdsourced annotation could motivate one, but it would be a new problem inspired by the paper, not a mirror of its 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.