[hotfix] Quantization patch; fix semantic_search_faiss/semantic_sea…

…rch_usearch rescoring (#2558)

* Correctly update the scores after rescoring

* Sort scores from high to low in recommended example

examples/applications/embedding-quantization/semantic_search_recommended.py

@@ -93,7 +93,7 @@ def search(query, top_k: int = 10, rescore_multiplier: int = 4):
 
     # 6. Sort the scores and return the top_k
     start_time = time.time()
-    indices = scores.argsort()[:top_k]
+    indices = (-scores).argsort()[:top_k]
     top_k_indices = binary_ids[indices]
     top_k_scores = scores[indices]
     sort_time = time.time() - start_time

sentence_transformers/quantization.py

@@ -140,7 +140,7 @@ def semantic_search_faiss(
         rescored_scores = np.einsum("ij,ikj->ik", rescore_embeddings, top_k_embeddings)
         rescored_indices = np.argsort(-rescored_scores)[:, :top_k]
         indices = indices[np.arange(len(query_embeddings))[:, None], rescored_indices]
-        scores = rescored_scores[:, :top_k]
+        scores = rescored_scores[np.arange(len(query_embeddings))[:, None], rescored_indices]
 
     delta_t = time.time() - start_t
 
@@ -293,7 +293,7 @@ def semantic_search_usearch(
         rescored_scores = np.einsum("ij,ikj->ik", rescore_embeddings, top_k_embeddings)
         rescored_indices = np.argsort(-rescored_scores)[:, :top_k]
         indices = indices[np.arange(len(query_embeddings))[:, None], rescored_indices]
-        scores = rescored_scores[:, :top_k]
+        scores = rescored_scores[np.arange(len(query_embeddings))[:, None], rescored_indices]
 
     delta_t = time.time() - start_t