| paper | d4-conf-sigecom-2010-0033 |
| 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 experiments and defines maximum social welfare as a benchmark, but it has no numbered theorem, lemma, corollary, or proposition asserting an algorithmic or computational result. The computational theory it mentions is cited background, so no qualifying result anchors a continuous population question.
fails bit a — no named computational result to mirror
None: the social-efficiency benchmark and behavioral findings are experimental; the balanced-outcome and matching claims are cited from prior work.
The paper contains no numbered Theorem, Lemma, Corollary, or Proposition asserting a computational result, so it has no qualifying anchor for a continuous mirror under your criteria.
The closest passages are Section 5.1’s comparison with maximum social welfare and Section 2’s account of the stable-and-balanced outcome result of Kleinberg and Tardos [13]. Neither is a named computational result proved in this paper: the former reports an experimental comparison, and the latter is cited background. I can’t honestly restate either as an anchor or build the requested prover-ready mirror around it.
The appended case offers no continuous-mirror anchor to defeat. It says the paper has no named computational result to mirror, which supports the negative case rather than the case for a mirror.
The strongest candidate in the paper is Section 5.1’s maximum-social-welfare benchmark: choose the largest set of deals subject to the network’s edges and each node’s deal limit. But the paper uses this optimization to normalize experimental performance; it does not present a computational result about the problem. The reported finding is that human subjects achieved roughly the greedy algorithm’s average efficiency, while leaving some feasible deals unclosed. Turning that benchmark into a high-multiplicity problem would require a society made of repeated network roles, with the links and deal limits preserved. That could define a continuous population model, but its question would be about centralized feasible deal selection, not the paper’s finding about how people negotiate.
The better version of a behavioral mirror would keep bargaining power, offers, deal limits, and network position in each agent’s type, then study a large population of such agents. Yet the paper’s behavioral findings concern outcomes of specific network structures and a specific interaction protocol: who is connected to whom, who proposes, who waits, and who refuses to close. Replacing those findings with computations over masses would require a new formal bargaining problem. Treating the continuum analytically would instead move toward the mean-field dynamics that the programme distinguishes from computational questions.
The other results do not supply a stronger anchor. The correlations between degree, deal limits, neighborhood structure, patience, and earnings are empirical findings, not complexity results. The stable-and-balanced outcome results are cited background, not results established by this paper. And the paper’s comparisons of human behavior with theory do not define an algorithmic task whose continuous-population complexity the paper studies.
So the strongest negative case is that this paper provides no computational result for ChoCo to continuize. A repeated-network bargaining or matching project might be worth pursuing on its own, but it would be a new project rather than a continuous mirror of this paper. That distinction is the weak point in a universal “no scenario” claim: a large repeated network is conceivable. The case against a *mirror of this paper* is much firmer than the case against every worthwhile continuous study of bargaining networks.
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.