diff --git a/CHANGELOG.markdown b/CHANGELOG.markdown
--- a/CHANGELOG.markdown
+++ b/CHANGELOG.markdown
@@ -1,3 +1,8 @@
+1.1.4
+------------
+* fix the problem with leading slashes for comments, drawings and pivot tables
+  (thanks to  mindston <mindston@eml.cc>)
+
 1.1.3
 ------------
 * fixed compatibility with xml-conduit moving `rsPretty` into internal module
diff --git a/data/absolute_relationships.xlsx b/data/absolute_relationships.xlsx
new file mode 100644
Binary files /dev/null and b/data/absolute_relationships.xlsx differ
diff --git a/src/Codec/Xlsx/Parser.hs b/src/Codec/Xlsx/Parser.hs
--- a/src/Codec/Xlsx/Parser.hs
+++ b/src/Codec/Xlsx/Parser.hs
@@ -214,8 +214,8 @@
 
       let commentsType = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/comments"
           commentTarget :: Maybe FilePath
-          commentTarget = relTarget <$> findRelByType commentsType sheetRels
-          legacyDrPath = fmap relTarget . flip Relationships.lookup sheetRels =<< legacyDrRId
+          commentTarget = logicalNameToZipItemName . relTarget <$> findRelByType commentsType sheetRels
+          legacyDrPath = fmap (logicalNameToZipItemName . relTarget) . flip Relationships.lookup sheetRels =<< legacyDrRId
       commentsMap <-
         fmap join . forM commentTarget $ getComments ar legacyDrPath
       let commentCells =
@@ -230,11 +230,11 @@
         fp <- lookupRelPath filePath sheetRels rId
         getTable ar fp
       drawing <- forM drawingRId $ \dId -> do
-        rel <- note (InvalidRef filePath dId) $ Relationships.lookup dId sheetRels
-        getDrawing ar contentTypes (relTarget rel)
+        fp <- lookupRelPath filePath sheetRels dId
+        getDrawing ar contentTypes fp
       let ptType = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/pivotTable"
       pivotTables <- forM (allByType ptType sheetRels) $ \rel -> do
-        let ptPath = relTarget rel
+        let ptPath = logicalNameToZipItemName $ relTarget rel
         bs <- note (MissingFile ptPath) $ Zip.fromEntry <$> Zip.findEntryByPath ptPath ar
         note (InconsistentXlsx $ "Bad pivot table in " <> T.pack ptPath) $
           parsePivotTable (flip Prelude.lookup caches) bs
@@ -369,9 +369,9 @@
 
   let commentsType = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/comments"
       commentTarget :: Maybe FilePath
-      commentTarget = relTarget <$> findRelByType commentsType sheetRels
+      commentTarget = logicalNameToZipItemName . relTarget <$> findRelByType commentsType sheetRels
       legacyDrRId = cur $/ element (n_ "legacyDrawing") >=> fromAttribute (odr"id")
-      legacyDrPath = fmap relTarget . flip Relationships.lookup sheetRels  =<< listToMaybe legacyDrRId
+      legacyDrPath = fmap (logicalNameToZipItemName . relTarget) . flip Relationships.lookup sheetRels  =<< listToMaybe legacyDrRId
 
   commentsMap :: Maybe CommentTable <- maybe (Right Nothing) (getComments ar legacyDrPath) commentTarget
 
@@ -470,14 +470,14 @@
 
   mDrawing <- case mDrawingId of
       Just dId -> do
-          rel <- note (InvalidRef filePath dId) $ Relationships.lookup dId sheetRels
-          Just <$> getDrawing ar contentTypes (relTarget rel)
+          fp <- lookupRelPath filePath sheetRels dId
+          Just <$> getDrawing ar contentTypes fp
       Nothing  ->
           return Nothing
 
   let ptType = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/pivotTable"
   pTables <- forM (allByType ptType sheetRels) $ \rel -> do
-    let ptPath = relTarget rel
+    let ptPath = logicalNameToZipItemName $ relTarget rel
     bs <- note (MissingFile ptPath) $ Zip.fromEntry <$> Zip.findEntryByPath ptPath ar
     note (InconsistentXlsx $ "Bad pivot table in " <> T.pack ptPath) $
       parsePivotTable (flip Prelude.lookup caches) bs
diff --git a/test/Main.hs b/test/Main.hs
--- a/test/Main.hs
+++ b/test/Main.hs
@@ -81,6 +81,10 @@
         $ floatsParsingTests toXlsx
     , testCase "toXlsxFast: correct floats parsing (typed and untyped cells are floats by default)"
         $ floatsParsingTests toXlsxFast
+    , testCase "toXlsx: correct resolving of absolute relationship targets"
+        $ absoluteRelationshipsTest toXlsx
+    , testCase "toXlsxFast: correct resolving of absolute relationship targets"
+        $ absoluteRelationshipsTest toXlsxFast
     , testGroup "Codec: sheet state visibility"
         [ testGroup "toXlsxEitherFast"
             [ testProperty "pure state == toXlsxEitherFast (fromXlsx (defXlsxWithState state))" $
@@ -123,6 +127,17 @@
         , ((4,1), def & cellValue ?~ CellDouble 15.0)
         ]
   expectedCells @==? parsedCells
+
+-- Test whether absolute logical item names are correctly mapped to ZIP item names.
+-- absolute_relationships.xlsx contains relationships of worksheet, comment,
+-- drawing and pivot table types with absolute paths in their targets.
+absoluteRelationshipsTest :: (ByteString -> Xlsx) -> IO ()
+absoluteRelationshipsTest parser = do
+  bs <- LB.readFile "data/absolute_relationships.xlsx"
+  let xlsx = parser bs
+      expectedComment = Just $ Comment (XlsxText "test comment") "author" False
+      parsedComment = xlsx ^? ixSheet "Test" . ixCell (2, 1) . cellComment . _Just
+  expectedComment @==? parsedComment
 
 constSheetName :: Text
 constSheetName = "sheet1"
diff --git a/xlsx.cabal b/xlsx.cabal
--- a/xlsx.cabal
+++ b/xlsx.cabal
@@ -1,6 +1,6 @@
 Name:                xlsx
 
-Version:             1.1.3
+Version:             1.1.4
 
 Synopsis:            Simple and incomplete Excel file parser/writer
 Description:
@@ -17,6 +17,7 @@
                      CHANGELOG.markdown
                      data/inline-strings.xlsx
                      data/floats.xlsx
+                     data/absolute_relationships.xlsx
 
 Homepage:            https://github.com/qrilka/xlsx
 Bug-Reports:         https://github.com/qrilka/xlsx/issues