Answers are grounded in this paper and cite their sources.
Data-asset attribution and posted-price revenue are related but distinct decisions. We specify a conditional cooperative-allocation architecture and report a six-asset pricing benchmark built from the red Wine Quality dataset. A fixed table of 64 cross-validated ridge-regression utilities supplies valuation anchors; simulated myopic buyers then purchase missing assets according to noisy marginal utilities. Across ten paired streams of 2,000 arrivals, the reported TrPS policy has a fixed-grid hindsight revenue difference of 354.9 ± 11.4 price units, versus 415.3 ± 12.4 for UCB1 and 363.2 ± 12.0 for static Shapley pricing. Removing its purchase-rate discount increases that difference to 369.1 ± 15.1; this tests a demand heuristic, not coalition stability. Variance-targeted sampling has no detectable revenue advantage in the reported summary. Truncated anchors have an efficiency excess of approximately 0.113, whose mechanism requires the original normalization and stopping records. Core projection, adaptive utility retraining and strategic seller behavior are not evaluated. The results describe one simulator and a restricted hindsight comparator; original code and seed-level records remain necessary to verify sampling, pricing and inference.
No takes yet. Share an insight, caveat, or question.
Behera et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: