| paper | d4-conf-atal-2003-0253 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
The numbered propositions characterize dialogue outcomes and tactical dependence, but none states an algorithm, complexity result, or computational formulation used to compute. A population-level acceptance or protocol-design objective would need to be added, so no anchor satisfies both requirements for a mirror.
fails bit a — no named computational result to mirror
No mirror covers the paper's numbered results: Propositions 1–11 characterize argumentation outcomes and dialogue protocols; the cited complexity results come from separate work.
This paper contains no named computational result of the required kind. Propositions 1–11 characterize dialogue outcomes; none classifies the complexity of a computational problem as tractable, hard, FPT, or otherwise.
For example, Proposition 11 says that, for two agents under any protocol, their acceptance outcomes equal the acceptable joint knowledge outcomes if and only if the outcomes do not depend on the order of assertions. Proposition 8 gives dialogues under the paper’s protocols whose acceptance outcomes do depend on assertion order. These are formal results, but they are not computational-complexity anchors. The paper’s references to complexity concern separate work, not results proved here.
A continuous population of agents grouped by complete knowledge-base and attitude types could be imagined, with mass representing each type’s share. But the paper studies two-agent dialogues, and turning that into a population-level problem would require adding a pairing model and a new computational objective. That would be a proposed extension, not a continuous mirror anchored in this paper’s named computational results.
I would reject a ChoCo mirror of this paper. Its strongest problem is that it proves no computational result to continuize: Propositions 1–11 characterize outcomes and their dependence on dialogue moves, while the complexity results are cited from separate work. That rules out a direct complexity mirror.
Propositions 8 and 11 are the best starting points for a broader proposal. One could distribute receiver types by mass, replay a dialogue with each type, and ask what share accepts a proposition under different assertion orders. Or one could seek a protocol that makes outcomes predetermined for most of that audience. These are coherent questions, but they add the population-level objective and computational problem the paper does not supply. For a fixed plan, the audience acceptance rate is just a weighted sum of per-type outcomes. To get a substantial algorithmic problem, one must add a shared protocol-design objective, a cost or disclosure budget, and a requirement across the audience. That would be interesting protocol synthesis or campaign optimization, but it is no longer a continuous mirror of the paper’s results.
The strongest scenario for a population is a service agent holding repeated persuasion dialogues with many recipients. Yet identical receiver types produce identical dialogues, so multiplicity only weights their outcomes; individualized knowledge bases can instead make the support grow with the audience. And the paper studies each bilateral exchange, not how a population aggregates acceptance into a collective decision. A distribution over recipients cannot by itself supply that missing rule.
So the case is strong against treating this paper as a ChoCo mirror, but it is not proof that no worthwhile extension exists. A mass-based protocol-design problem could be valuable on its own. The distinction is that its computational core would have to be built around new population and optimization choices, rather than extracted from a computational result in this paper.
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.