diff --git a/tests/test_frequency.py b/tests/test_frequency.py
new file mode 100644
index 0000000000000..680e535d3570b
--- /dev/null
+++ b/tests/test_frequency.py
@@ -0,0 +1,42 @@
+import tempfile
+import unittest
+from pathlib import Path
+
+from trendradar.core.frequency import load_frequency_words, matches_word_groups
+
+
+class GlobalFilterTests(unittest.TestCase):
+    def test_global_filter_supports_regular_expressions(self):
+        config = """[GLOBAL_FILTER]
+/spam|clickbait/
+
+[WORD_GROUPS]
+news
+"""
+
+        with tempfile.TemporaryDirectory() as temp_dir:
+            config_path = Path(temp_dir) / "frequency_words.txt"
+            config_path.write_text(config, encoding="utf-8")
+            word_groups, filter_words, global_filters = load_frequency_words(
+                str(config_path)
+            )
+
+        self.assertFalse(
+            matches_word_groups(
+                "Breaking news: spam", word_groups, filter_words, global_filters
+            )
+        )
+        self.assertFalse(
+            matches_word_groups(
+                "Clickbait news", word_groups, filter_words, global_filters
+            )
+        )
+        self.assertTrue(
+            matches_word_groups(
+                "Breaking news", word_groups, filter_words, global_filters
+            )
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/trendradar/core/frequency.py b/trendradar/core/frequency.py
index 561741ac28455..7fe3a2c9793dc 100644
--- a/trendradar/core/frequency.py
+++ b/trendradar/core/frequency.py
@@ -95,7 +95,7 @@ def _word_matches(word_config: Union[str, Dict], title_lower: str) -> bool:
 
 def load_frequency_words(
     frequency_file: Optional[str] = None,
-) -> Tuple[List[Dict], List[str], List[str]]:
+) -> Tuple[List[Dict], List[Dict], List[Dict]]:
     """
     加载频率词配置
 
@@ -161,13 +161,13 @@ def load_frequency_words(
 
         # 处理全局过滤区域
         if current_section == "GLOBAL_FILTER":
-            # 直接添加所有非空行到全局过滤列表
+            # 解析所有非空行，支持普通词和正则语法
             for line in lines:
                 # 忽略特殊语法前缀，只提取纯文本
                 if line.startswith(("!", "+", "@")):
                     continue  # 全局过滤区不支持特殊语法
                 if line:
-                    global_filters.append(line)
+                    global_filters.append(_parse_word(line))
             continue
 
         # 处理词组区域
@@ -247,7 +247,7 @@ def matches_word_groups(
     title: str,
     word_groups: List[Dict],
     filter_words: List,
-    global_filters: Optional[List[str]] = None
+    global_filters: Optional[List[Union[str, Dict]]] = None,
 ) -> bool:
     """
     检查标题是否匹配词组规则
@@ -271,7 +271,9 @@ def matches_word_groups(
 
     # 全局过滤检查（优先级最高）
     if global_filters:
-        if any(global_word.lower() in title_lower for global_word in global_filters):
+        if any(
+            _word_matches(global_item, title_lower) for global_item in global_filters
+        ):
             return False
 
     # 如果没有配置词组，则匹配所有标题（支持显示全部新闻）