| paper | d4-conf-ijcai-1997-0017 |
| 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 develops a semantics and constructive characterization, but contains no numbered theorem, lemma, corollary, or proposition asserting an algorithm, complexity bound, or hardness result; it explicitly leaves algorithms for future work [paper](https://www.ijcai.org/Proceedings/97-1/Papers/027.pdf). A population-of-cases task would also require an aggregation semantics the paper does not supply, but that is secondary because bit (a) fails.
fails bit a — no named computational result to mirror
No paper result is covered by a continuous-population mirror.
I can’t make an honest anchored case for a continuous mirror here. The paper’s contribution is a semantics for dynamic preferences on assumptions, an equivalent constructive characterization, and the claim that the resulting inference relation inherits Lehmann’s System C properties. Those are semantic results, not numbered results classifying the complexity of a computational problem. The conclusion even says that algorithms and applications are future work. [Junker, “A Cumulative-Model Semantics for Dynamic Preferences on Assumptions”](https://www.ijcai.org/Proceedings/97-1/Papers/027.pdf)
So there is no eligible anchor to mirror under your rules. A continuous population of reasoners or assumption types could motivate a new question, but this paper supplies no named computational result for that question to continue.
The proponent raises no eligible anchors, and that is the strongest case against a continuous mirror here. The paper’s results concern the semantics of dynamic preferences, a constructive characterization, and preservation of properties of Lehmann’s System C. It gives no named complexity result for a computational problem. “Constructive” does not supply a running-time claim, and the conclusion explicitly leaves algorithms and applications for future work.
The best candidate for a mirror would be a new task such as computing preferred assumption sets or testing whether a conclusion follows. One could imagine many reasoning cases drawn from a distribution of types. But if the cases share the same complete theory and preferences, they are interchangeable copies of one inference problem. If they differ, computing an aggregate answer requires a population-level aggregation semantics that this paper does not define. Either way, the proposed task would be new computational work built on the paper’s semantics, not a continuous version of a result the paper establishes.
That makes the negative case strong under the programme’s requirement for an anchored mirror. It does not prove that no useful continuous-population question could ever be developed in this area; future algorithmic work or a new collective-reasoning model might be worthwhile. But neither would be a mirror of this paper’s contribution as presented.
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.