diff --git a/README.md b/README.md
index dfce365..1857813 100644
--- a/README.md
+++ b/README.md
@@ -1,9 +1,13 @@
 # CJ Unpolarized DIS Database
-World __proton__ and __deuteron__ data of unploarized cross sections, structure function F2, and the longitudinal to transverse cross section ratio R are collected or extracted from various experiments. Data were collected for the CJ global fit and related analysis. Now open for general use. 
+World __proton__ and __deuteron__ data of unploarized DIS cross sections, F2 structure functions, and the longitudinal to transverse cross section ratio R are collected or extracted from various experiments. Data were collected for the CJ global fit and related analysis. Now open for general use. 
 
-__Reference__: appendix A in xxx. See also [note][note] for reduced cross section and F2 calculation.
+__Reference__: appendix A in [arXiv:2309.16851][https://arxiv.org/abs/2309.16851]. See also [note][note] for reduced cross section and F2 calculation.
 
-Also available: combined sets of normalized and rebinned F2p and F2d world data, as well as the extracted F2n and F2n/F2p with CJ15 analysis from xxx. 
+Also available: combined sets of normalized F2p and F2d world data, as well as the extracted F2n (from p and d data) and F2n/F2p (from d/p data) with CJ15 analysis (see arXiv:2309.16851). Each table gives F2 in rebinned and original Q2.
+[F2 p][data/rebin_p.xlsx]
+[F2 d][data/rebin_d.xlsx]
+[F2 n][data/rebin_n.xlsx]
+[F2 n/p][data/f2np_all.xlsx]
 
 ## Data table
 Each set is provided in both plain text and xlsx/csv format. Filenames and their ID (100xx) in database are listed on the table below. Click on the names for details. Data format is explained on the [wikipage](https://github.com/JeffersonLab/CJ-database/wiki).
diff --git a/data/f2np_all.xlsx b/data/f2np_all.xlsx
new file mode 100644
index 0000000..516f1e7
Binary files /dev/null and b/data/f2np_all.xlsx differ
diff --git a/data/rebin_d.xlsx b/data/rebin_d.xlsx
new file mode 100644
index 0000000..21a9d6e
Binary files /dev/null and b/data/rebin_d.xlsx differ
diff --git a/data/rebin_n.xlsx b/data/rebin_n.xlsx
new file mode 100644
index 0000000..22f4042
Binary files /dev/null and b/data/rebin_n.xlsx differ
diff --git a/data/rebin_p.xlsx b/data/rebin_p.xlsx
new file mode 100644
index 0000000..31e3651
Binary files /dev/null and b/data/rebin_p.xlsx differ
diff --git a/tocsv.py b/tocsv.py
new file mode 100644
index 0000000..824e391
--- /dev/null
+++ b/tocsv.py
@@ -0,0 +1,26 @@
+import pandas as pd
+import sys
+import os
+if len(sys.argv)>1:
+	excel_files = [sys.argv[1]]
+else: 
+	# excel_files = ["100%02d.xlsx" %(ii+1) for ii in range(70)]
+    excel_files =os.listdir('data/JAM/')
+
+n = 0
+for excel_file in excel_files:
+    if not ('100' in excel_file):
+        continue
+# for excel_file in excel_files:
+    # if excel_file=="10012.xlsx" or excel_file=="10013.xlsx" or excel_file=="10023.xlsx" or excel_file=="10024.xlsx":
+        # continue
+    print("Converting '{}'".format(excel_file))
+    # try:
+    #     df = pd.read_excel("format/"+excel_file,"format")
+    #     output = "csv/"+excel_file.split('.')[0]+'.csv'
+    #     df.to_csv(output)    
+    # except:
+    #     print("  Failed to convert")
+    df = pd.read_excel("data/JAM/"+excel_file,"format")
+    output = "data/JAM/csv/"+excel_file.split('.')[0]+'.csv'
+    df.to_csv(output)