diff --git a/internal/core/src/segcore/reduce_c.cpp b/internal/core/src/segcore/reduce_c.cpp
index c84bf29697..af13ede75b 100644
--- a/internal/core/src/segcore/reduce_c.cpp
+++ b/internal/core/src/segcore/reduce_c.cpp
@@ -64,6 +64,7 @@ GetResultData(std::vector<std::vector<int64_t>>& search_records,
     auto num_segments = search_results.size();
     AssertInfo(num_segments > 0, "num segment must greater than 0");
 
+    std::unordered_set<int64_t> pk_set;
     int64_t skip_dup_cnt = 0;
     for (int64_t qi = 0; qi < nq; qi++) {
         std::vector<SearchResultPair> result_pairs;
@@ -86,38 +87,25 @@ GetResultData(std::vector<std::vector<int64_t>>& search_records,
             search_records[index].push_back(result_pair.offset_++);
         }
 #else
-        float prev_dis = MAXFLOAT;
-        std::unordered_set<int64_t> prev_pk_set;
+        pk_set.clear();
         while (curr_offset - base_offset < topk) {
             result_pairs[0].reset_distance();
             std::sort(result_pairs.begin(), result_pairs.end(), std::greater<>());
             auto& result_pair = result_pairs[0];
             auto index = result_pair.index_;
             int64_t curr_pk = result_pair.search_result_->primary_keys_[result_pair.offset_];
-            float curr_dis = result_pair.search_result_->result_distances_[result_pair.offset_];
             // remove duplicates
-            if (curr_pk == INVALID_ID || std::abs(curr_dis - prev_dis) > 0.00001) {
+            if (curr_pk == INVALID_ID || pk_set.count(curr_pk) == 0) {
                 result_pair.search_result_->result_offsets_.push_back(curr_offset++);
-                search_records[index].push_back(result_pair.offset_);
-                prev_dis = curr_dis;
-                prev_pk_set.clear();
-                prev_pk_set.insert(curr_pk);
-            } else {
-                // To handle this case:
-                //    e1: [100, 0.99]
-                //    e2: [101, 0.99]   ==> not duplicated, should keep
-                //    e3: [100, 0.99]   ==> duplicated, should remove
-                if (prev_pk_set.count(curr_pk) == 0) {
-                    result_pair.search_result_->result_offsets_.push_back(curr_offset++);
-                    search_records[index].push_back(result_pair.offset_);
-                    // prev_pk_set keeps all primary keys with same distance
-                    prev_pk_set.insert(curr_pk);
-                } else {
-                    // the entity with same distance and same primary key must be duplicated
-                    skip_dup_cnt++;
+                search_records[index].push_back(result_pair.offset_++);
+                if (curr_pk != INVALID_ID) {
+                    pk_set.insert(curr_pk);
                 }
+            } else {
+                // skip entity with same primary key
+                result_pair.offset_++;
+                skip_dup_cnt++;
             }
-            result_pair.offset_++;
         }
 #endif
     }
diff --git a/internal/proxy/task.go b/internal/proxy/task.go
index bb8a2f2179..26090d5e59 100644
--- a/internal/proxy/task.go
+++ b/internal/proxy/task.go
@@ -1812,8 +1812,7 @@ func reduceSearchResultData(searchResultData []*schemapb.SearchResultData, nq in
 	for i := int64(0); i < nq; i++ {
 		offsets := make([]int64, len(searchResultData))
 
-		var prevIDSet = make(map[int64]struct{})
-		var prevScore float32 = math.MaxFloat32
+		var idSet = make(map[int64]struct{})
 		var j int64
 		for j = 0; j < topk; {
 			sel := selectSearchResultData(searchResultData, offsets, topk, i)
@@ -1830,28 +1829,15 @@ func reduceSearchResultData(searchResultData []*schemapb.SearchResultData, nq in
 			}
 
 			// remove duplicates
-			if math.Abs(float64(score)-float64(prevScore)) > 0.00001 {
+			if _, ok := idSet[id]; !ok {
 				typeutil.AppendFieldData(ret.Results.FieldsData, searchResultData[sel].FieldsData, idx)
 				ret.Results.Ids.GetIntId().Data = append(ret.Results.Ids.GetIntId().Data, id)
 				ret.Results.Scores = append(ret.Results.Scores, score)
-				prevScore = score
-				prevIDSet = map[int64]struct{}{id: {}}
+				idSet[id] = struct{}{}
 				j++
 			} else {
-				// To handle this case:
-				//    e1: [100, 0.99]
-				//    e2: [101, 0.99]   ==> not duplicated, should keep
-				//    e3: [100, 0.99]   ==> duplicated, should remove
-				if _, ok := prevIDSet[id]; !ok {
-					typeutil.AppendFieldData(ret.Results.FieldsData, searchResultData[sel].FieldsData, idx)
-					ret.Results.Ids.GetIntId().Data = append(ret.Results.Ids.GetIntId().Data, id)
-					ret.Results.Scores = append(ret.Results.Scores, score)
-					prevIDSet[id] = struct{}{}
-					j++
-				} else {
-					// entity with same id and same score must be duplicated
-					skipDupCnt++
-				}
+				// skip entity with same id
+				skipDupCnt++
 			}
 			offsets[sel]++
 		}