diff --git a/chrome/content/zotero/collectionViewItemTree.jsx b/chrome/content/zotero/collectionViewItemTree.jsx
index a55779a622..2d8612522b 100644
--- a/chrome/content/zotero/collectionViewItemTree.jsx
+++ b/chrome/content/zotero/collectionViewItemTree.jsx
@@ -566,7 +566,12 @@ class CollectionViewItemTreeRowProvider extends ItemTreeRowProvider {
 		this.itemTree._refreshPromise = deferred.promise;
 		
 		try {
-			this.collectionTreeRows.forEach(row => row.clearCache());
+			// A best-match rerank after an embeddings change reuses the cached
+			// search results -- only the ranking depends on the embeddings, so
+			// there's no need to re-run the underlying search
+			if (!options.reuseSearchResults) {
+				this.collectionTreeRows.forEach(row => row.clearCache());
+			}
 			this._bestMatchRanks = null;
 			this._bestMatchBarFractions = null;
 			this._bestMatchIndexState = null;
@@ -880,6 +885,7 @@ class CollectionViewItemTreeRowProvider extends ItemTreeRowProvider {
 
 		var madeChanges = false;
 		var refresh = false;
+		var reuseSearchResults = false;
 		var sort = false;
 
 		// Selection strategy
@@ -933,7 +939,15 @@ class CollectionViewItemTreeRowProvider extends ItemTreeRowProvider {
 		if (action == 'refresh' && type == 'item' && extraData && extraData.embeddingsUpdate
 				&& collectionTreeRows.some(rowIsBestMatchSearch)) {
 			// The background indexer committed new or changed embeddings, so
-			// rerun the active best-match search to update the scores and ranks
+			// rerun the active best-match search to update the scores and ranks.
+			// Only the ranking depends on the embeddings, so unless a selected
+			// row trims membership by score (a top-K cutoff), the underlying
+			// search results are unchanged and can be reused while just the
+			// ranking is recomputed.
+			reuseSearchResults = collectionTreeRows.every((row) => {
+				return typeof row.hasBestMatchCutoff == 'function'
+					&& !row.hasBestMatchCutoff();
+			});
 			this.itemTree.invalidateRowCache(ids);
 			refresh = true;
 			madeChanges = true;
@@ -1213,7 +1227,7 @@ class CollectionViewItemTreeRowProvider extends ItemTreeRowProvider {
 		}
 
 		if (refresh) {
-			await this._refresh();
+			await this._refresh({ reuseSearchResults });
 		}
 		if (sort) {
 			await this.itemTree._ensureSortContextReady();
diff --git a/chrome/content/zotero/xpcom/collectionTreeRow.js b/chrome/content/zotero/xpcom/collectionTreeRow.js
index e67ffe9b67..23bdebe08a 100644
--- a/chrome/content/zotero/xpcom/collectionTreeRow.js
+++ b/chrome/content/zotero/xpcom/collectionTreeRow.js
@@ -759,6 +759,29 @@ Zotero.CollectionTreeRow.prototype.getBestMatchQuery = function () {
 	return source ? source.getBestMatchQuery().query : this.searchText;
 };
 
+/**
+ * Whether an active best-match search has a top-K cutoff, which trims
+ * membership by score and so makes the underlying
+ * Zotero.Search.prototype.search() itself depend on the embeddings. Without
+ * one, the search results are stable across embedding changes, so they can be
+ * reused while only the ranking is recomputed.
+ *
+ * @return {Boolean}
+ */
+Zotero.CollectionTreeRow.prototype.hasBestMatchCutoff = function () {
+	let source = this.getBestMatchSource();
+	if (source) {
+		return !!source.getBestMatchQuery().topK;
+	}
+	// A quick-search best-match doesn't trim membership itself, but a selected
+	// saved search may still carry its own top-K cutoff
+	if (this.isSearch() && this.ref instanceof Zotero.Search) {
+		let query = this.ref.getBestMatchQuery();
+		return !!(query && query.topK);
+	}
+	return false;
+};
+
 Zotero.CollectionTreeRow.prototype.isSortable = function () {
 	return !this.isFeedsOrFeed() && !this.isRecentlyRead();
 }
diff --git a/test/tests/collectionViewItemTreeTest.js b/test/tests/collectionViewItemTreeTest.js
index 2358e993c7..225c6798ec 100644
--- a/test/tests/collectionViewItemTreeTest.js
+++ b/test/tests/collectionViewItemTreeTest.js
@@ -555,6 +555,32 @@ describe("CollectionViewItemTree", function () {
 				await itemsView._refreshPromise;
 				assert.deepEqual(itemsView._rows.map(row => row.id), [itemA.id, itemB.id]);
 			});
+
+			it("should rerank without re-running the search when embeddings change", async function () {
+				let col = await createDataObject('collection');
+				let itemA = await createDataObject('item', { title: "reuse A", collections: [col.id] });
+				let itemB = await createDataObject('item', { title: "reuse B", collections: [col.id] });
+				let best = itemA.id;
+				stubs.push(sinon.stub(Zotero.Embeddings, 'scoreItemIDs').callsFake(
+					async (query, itemIDs) => new Map(itemIDs.map(id => [id, id == best ? 0.9 : 0.5]))
+				));
+
+				await select(win, col);
+				itemsView = zp.itemsView;
+				await itemsView.setFilter('search', 'some query');
+				assert.deepEqual(itemsView._rows.map(row => row.id), [itemA.id, itemB.id]);
+
+				// The underlying search is dropped and re-run by clearing the row cache
+				let clearCacheSpy = sinon.spy(Zotero.CollectionTreeRow.prototype, 'clearCache');
+				stubs.push(clearCacheSpy);
+
+				best = itemB.id;
+				await Zotero.Notifier.trigger('refresh', 'item', [itemA.id, itemB.id], { embeddingsUpdate: true });
+				await itemsView._refreshPromise;
+
+				assert.deepEqual(itemsView._rows.map(row => row.id), [itemB.id, itemA.id]);
+				assert.isFalse(clearCacheSpy.called);
+			});
 		});
 
 		it("should expand parent item and attachment for an annotation match", async function () {