feat: update knowledge base retrieval configuration and UI adjustments

astrbot/core/config/default.py

@@ -137,7 +137,8 @@ DEFAULT_CONFIG = {
     "default_kb_collection": "",  # 默认知识库名称, 已经过时
     "plugin_set": ["*"],  # "*" 表示使用所有可用的插件, 空列表表示不使用任何插件
     "kb_names": [],  # 默认知识库名称列表
-    "kb_final_top_k": 5,  # 知识库检索的最终返回结果数量
+    "kb_fusion_top_k": 20,  # 知识库检索融合阶段返回结果数量
+    "kb_final_top_k": 5,  # 知识库检索最终返回结果数量
 }
 
 
@@ -2003,6 +2004,7 @@ CONFIG_METADATA_2 = {
                 "type": "string",
             },
             "kb_names": {"type": "list", "items": {"type": "string"}},
+            "kb_fusion_top_k": {"type": "int", "default": 20},
             "kb_final_top_k": {"type": "int", "default": 5},
         },
     },
@@ -2089,6 +2091,11 @@ CONFIG_METADATA_3 = {
                         "_special": "select_knowledgebase",
                         "hint": "支持多选",
                     },
+                    "kb_fusion_top_k": {
+                        "description": "融合检索结果数",
+                        "type": "int",
+                        "hint": "多个知识库检索结果融合后的返回结果数量",
+                    },
                     "kb_final_top_k": {
                         "description": "最终返回结果数",
                         "type": "int",

astrbot/core/knowledge_base/kb_mgr.py

@@ -212,6 +212,7 @@ class KnowledgeBaseManager:
         self,
         query: str,
         kb_names: list[str],
+        top_k_fusion: int = 20,
         top_m_final: int = 5,
     ) -> dict | None:
         """从指定知识库中检索相关内容"""
@@ -229,6 +230,7 @@ class KnowledgeBaseManager:
             query=query,
             kb_ids=kb_ids,
             kb_id_helper_map=kb_id_helper_map,
+            top_k_fusion=top_k_fusion,
             top_m_final=top_m_final,
         )
         if not results:

astrbot/core/knowledge_base/retrieval/manager.py

@@ -61,6 +61,7 @@ class RetrievalManager:
         query: str,
         kb_ids: List[str],
         kb_id_helper_map: dict[str, KBHelper],
+        top_k_fusion: int = 20,
         top_m_final: int = 5,
     ) -> List[RetrievalResult]:
         """混合检索
@@ -120,7 +121,7 @@ class RetrievalManager:
         fused_results = await self.rank_fusion.fuse(
             dense_results=dense_results,
             sparse_results=sparse_results,
-            top_k=kb_options.get("top_k_fusion", 20),
+            top_k=top_k_fusion,
         )
 
         # 4. 转换为 RetrievalResult (获取元数据)
@@ -209,7 +210,8 @@ class RetrievalManager:
 
         # 按相似度排序并返回 top_k
         all_results.sort(key=lambda x: x.similarity, reverse=True)
-        return all_results[: len(all_results) // len(kb_ids)]
+        # return all_results[: len(all_results) // len(kb_ids)]
+        return all_results
 
     async def _rerank(
         self,

astrbot/core/knowledge_base/retrieval/sparse_retriever.py

@@ -124,4 +124,5 @@ class SparseRetriever:
             )
 
         results.sort(key=lambda x: x.score, reverse=True)
-        return results[: len(results) // len(kb_ids)]
+        # return results[: len(results) // len(kb_ids)]
+        return results

astrbot/core/pipeline/process_stage/utils.py

@@ -16,6 +16,7 @@ async def inject_kb_context(
     """
     kb_mgr = p_ctx.plugin_manager.context.kb_manager
     kb_names = p_ctx.astrbot_config.get("kb_names", [])
+    top_k_fusion = p_ctx.astrbot_config.get("kb_fusion_top_k", 20)
     top_k = p_ctx.astrbot_config.get("kb_final_top_k", 5)
 
     if not kb_names:
@@ -24,6 +25,7 @@ async def inject_kb_context(
     kb_context = await kb_mgr.retrieve(
         query=req.prompt,
         kb_names=kb_names,
+        top_k_fusion=top_k_fusion,
         top_m_final=top_k,
     )
     if not kb_context:

dashboard/src/views/knowledge-base/components/SettingsTab.vue

@@ -34,7 +34,7 @@
           <h3 class="text-h6 mb-4 mt-6">{{ t('settings.retrieval') }}</h3>
 
           <v-row>
-            <v-col cols="12" md="4">
+            <v-col cols="12" md="6">
               <v-text-field
                 v-model.number="formData.top_k_dense"
                 :label="t('settings.topKDense')"
@@ -43,7 +43,7 @@
                 density="comfortable"
               />
             </v-col>
-            <v-col cols="12" md="4">
+            <v-col cols="12" md="6">
               <v-text-field
                 v-model.number="formData.top_k_sparse"
                 :label="t('settings.topKSparse')"
@@ -52,7 +52,7 @@
                 density="comfortable"
               />
             </v-col>
-            <v-col cols="12" md="4">
+            <!-- <v-col cols="12" md="4">
               <v-text-field
                 v-model.number="formData.top_m_final"
                 :label="t('settings.topMFinal')"
@@ -60,7 +60,7 @@
                 variant="outlined"
                 density="comfortable"
               />
-            </v-col>
+            </v-col> -->
           </v-row>
 
           <!-- 模型设置 -->
@@ -213,7 +213,7 @@ watch(() => props.kb, (kb) => {
       chunk_overlap: kb.chunk_overlap || 50,
       top_k_dense: kb.top_k_dense || 50,
       top_k_sparse: kb.top_k_sparse || 50,
-      top_m_final: kb.top_m_final || 5,
+      // top_m_final: kb.top_m_final || 5,
       embedding_provider_id: kb.embedding_provider_id || '',
       rerank_provider_id: kb.rerank_provider_id || ''
     }
@@ -281,7 +281,7 @@ const saveSettings = async () => {
       chunk_overlap: formData.value.chunk_overlap,
       top_k_dense: formData.value.top_k_dense,
       top_k_sparse: formData.value.top_k_sparse,
-      top_m_final: formData.value.top_m_final,
+      // top_m_final: formData.value.top_m_final,
       rerank_provider_id: formData.value.rerank_provider_id
     })