diff --git a/deepdoc/parser/pdf_parser.py b/deepdoc/parser/pdf_parser.py
index 5068eed62b..4e43df7a31 100644
--- a/deepdoc/parser/pdf_parser.py
+++ b/deepdoc/parser/pdf_parser.py
@@ -392,7 +392,7 @@ def _naive_vertical_merge(self):
                 b["text"].strip()[-1] in ",;:'\"，、‘“；：-",
                 len(b["text"].strip()) > 1 and b["text"].strip(
                 )[-2] in ",;:'\"，‘“、；：",
-                b["text"].strip()[0] in "。；？！?”）),，、：",
+                b_["text"].strip()[0] in "。；？！?”）),，、：",
             ]
             # features for not concating
             feats = [
diff --git a/rag/app/naive.py b/rag/app/naive.py
index f91734b27a..f1e3e3bdb2 100644
--- a/rag/app/naive.py
+++ b/rag/app/naive.py
@@ -19,6 +19,8 @@
 from rag.nlp import rag_tokenizer, naive_merge, tokenize_table, tokenize_chunks, find_codec
 from deepdoc.parser import PdfParser, ExcelParser, DocxParser
 from rag.settings import cron_logger
+from rag.utils import num_tokens_from_string
+
 
 class Docx(DocxParser):
     def __init__(self):
@@ -149,8 +151,14 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
                     if not l:
                         break
                     txt += l
-        sections = txt.split("\n")
-        sections = [(l, "") for l in sections if l]
+        sections = []
+        for sec in txt.split("\n"):
+            if num_tokens_from_string(sec) > 10 * parser_config.get("chunk_token_num", 128):
+                sections.append((sec[:int(len(sec)/2)], ""))
+                sections.append((sec[int(len(sec)/2):], ""))
+            else:
+                sections.append((sec, ""))
+
         callback(0.8, "Finish parsing.")
 
     elif re.search(r"\.doc$", filename, re.IGNORECASE):
@@ -163,7 +171,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
 
     else:
         raise NotImplementedError(
-            "file type not supported yet(doc, docx, pdf, txt supported)")
+            "file type not supported yet(pdf, xlsx, doc, docx, txt supported)")
 
     st = timer()
     chunks = naive_merge(
diff --git a/rag/nlp/rag_tokenizer.py b/rag/nlp/rag_tokenizer.py
index abd9146a94..be5b724b92 100644
--- a/rag/nlp/rag_tokenizer.py
+++ b/rag/nlp/rag_tokenizer.py
@@ -24,7 +24,7 @@ def rkey_(self, line):
     def loadDict_(self, fnm):
         print("[HUQIE]:Build trie", fnm, file=sys.stderr)
         try:
-            of = open(fnm, "r")
+            of = open(fnm, "r", encoding='utf-8')
             while True:
                 line = of.readline()
                 if not line:
diff --git a/requirements.txt b/requirements.txt
index f242a2ce3e..c8c3cd25d1 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -136,3 +136,4 @@ BCEmbedding
 loguru==0.7.2
 umap-learn
 fasttext==0.9.2
+volcengine
diff --git a/requirements_dev.txt b/requirements_dev.txt
index f29022bcac..ae0ba7a348 100644
--- a/requirements_dev.txt
+++ b/requirements_dev.txt
@@ -124,3 +124,4 @@ ollama==0.1.8
 redis==5.0.4
 fasttext==0.9.2
 umap-learn
+volcengine