OriginCount: Counting Origin Events Instead of Pages in Retrieval Agents

A retrieval agent answers a question by fetching pages, counting how many of them support a candidate answer, and stopping once the count is high enough. That count is usually a count of pages. On the web many pages restate one original record, a wire report, a press release, or a dataset card, without adding a new observation; we call such a page an echo of its origin. An echo carries no evidence beyond its origin: if pages $D_1,\dots ,D_m$ are produced from a record $E$ and add nothing to it, then $\mathrm{Pr}(H\mid E,D_{1:m})=\mathrm{Pr}(H\mid E)$, so a page count applies the evidence of $E$ once for $E$ and once more for every echo. On six web snapshots with page-level provenance labels, $34\%$ of retrieved pages are echoes, and counting by origin instead of by page changes the stopping round or the answer on $18\%$ of the questions. OriginCount groups pages by the origin they can be traced to, gives each group one vote, keeps a page of unknown origin as its own vote, and leaves the content of every page pooled. On a Qwen3-32B host it lowers the rate of confidently wrong answers from $16.4\%$ to $11.3\%$ and the calibration error from $0.150$ to $0.090$, at a cost of one point of accuracy, which the true labels trace to grouping errors, and a rule-based provenance extractor makes it cost $2$ ms per page; every comparison is paired by question at one budget of eight prefetched calls. The correction repairs copy inflation and nothing more: it cannot make distinct origins independent, and where copies had propped up a correct answer, grouping them lowers accuracy.